IRC log for #htc-linux on 20100227

00:00.14bzofinally, my acpuclock merge kernel does not crash!
00:00.52ToAsTcfh?
00:02.23Spirits-Sighttmzt: I have not found the thread that I should go ask?  do you know the site address for it?  This was about the bootloader image that shows 1inch of white at bottom, also the screen before I see the peguin is not readable and shows messed up look, sorry don't know how to describe it.
00:02.39tmztthe nand thread
00:03.07Spirits-Sightwhich one, I have found a few with nand in them
00:03.24Spirits-Sighthttp://forum.xda-developers.com/showthread.php?t=614275
00:05.57*** join/#htc-linux mastermerlin (~merlin@pD957B5A0.dip.t-dialin.net)
00:08.37Markinustmzt: which device do you have?
00:10.13tmztrhod500
00:10.25ToAsTcfhbzo ok i think i figured out my issue. its not wanting me to set the max in the config higher then the clock defined in acpuclock.. anyhow when i do it wont boot. but if i set the max lower then whats in acpulock then it will boot, but not move to the next pll. so whatever the issue is. its not liking to go to pll2
00:10.33NetRippercr2_, http://netripper.pastebin.com/a7bM48Sh doesn't really say what irq... 65535, so need to check where it comes from..
00:11.13NetRippercr2_, i suppose that's -1, some error somewhere?
00:11.39NetRipperand yes, i forgot a \n behind that fb console line :p
00:12.52NetRipperim off to bed as well
00:12.52NetRippernight :)
00:18.43*** join/#htc-linux ccube (~ccube@95-90-193-11-dynip.superkabel.de)
00:20.49ToAsTcfhbzo do u think setting the clock from the acpuclock.c may be the issue?
00:22.01bzono, should be fine
00:25.46ToAsTcfhhmmm
00:26.09ToAsTcfhits just not wanting to jump up for some reaso
00:26.12ToAsTcfhn
00:26.13Spirits-Sighttmzt: OK I have found this one also: http://forum.xda-developers.com/showthread.php?p=5504855  I don't know which one to ask on
00:26.38*** join/#htc-linux Zoolooc (~fredsibar@p54957D24.dip.t-dialin.net)
00:29.58ToAsTcfhphh: thats cool u add turbo clock tables to ur source
00:30.42ToAsTcfhbzo yeah idk it could be the whole turbo mode all together
00:31.28bzoyou could try turning the debug flags on acpuclock and see what dmesg says
00:32.18ToAsTcfheh yeah ive already donee that but when its not working im not getting an init at all
00:32.46*** join/#htc-linux balsat (~balsat@87.72.13.34)
00:33.01ToAsTcfhits not doing anything but hanging and  waiting
00:34.08*** join/#htc-linux mickey|bbl (~mickey@80.81.242.146)
00:37.19ToAsTcfhbzo : [    6.994567] msm_cpufreq_target 537600 r 1 (245760-537600) selected 537600
00:37.19ToAsTcfhNo errors detected
00:37.21ToAsTcfhlol
00:37.52ToAsTcfh547 is suppose to be the clock
00:37.59ToAsTcfh54720
00:39.08bzowhere are there any 537 or 547 clocks?
00:39.55ToAsTcfhnowhere
00:40.03ToAsTcfhlol thats what i mean
00:40.20ToAsTcfh547 is in the acpuclock.c
00:40.27ToAsTcfhand iin config
00:40.58ToAsTcfhbzo i wonder if i ccan just stay on pll1 and over clock it?
00:41.21ToAsTcfhi mean i have no though that thinks ill reach 1000000
00:41.38ToAsTcfh800 tops
00:42.30ToAsTcfhidk its def not liking pll2 being uped
00:43.59ToAsTcfh[    3.221130] OC: ADJUSTING FREQ TABLE freq=179200 div=2 ahbclk=89600 ahbdiv=1
00:43.59ToAsTcfh[    3.221160] OC: ADJUSTING FREQ TABLE freq=268800 div=1 ahbclk=67200 ahbdiv=3
00:43.59ToAsTcfh[    3.221191] OC: ADJUST freq_tbl_entry: 528000 to 537600
00:43.59ToAsTcfh[    3.221191] OC: ADJUSTING FREQ TABLE freq=537600 div=0 ahbclk=134400 ahbdiv=3
00:43.59ToAsTcfh[    3.221893] ACPU running at 480000 KHz
00:44.00ToAsTcfh[    3.232757] msm_i2c_probe
00:44.02ToAsTcfh[    3.233001] msm_i2c_probe: clk_ctl 35d, 100000 Hz
00:47.20ToAsTcfhbzo: CONFIG_CPU_FREQ_GOV_MSM7K=y?
00:49.28bzoprobably fine
00:52.43ToAsTcfhbzo are ur clocks defined in ur board file?
00:53.01bzono
00:53.12ToAsTcfhboard.heroc has its clocks defined
00:59.11ToAsTcfhbzo : in my board http://en.pastebin.ca/1812716
01:02.42ToAsTcfhcould that be th issue?
01:04.09bzodon't know, guess you could try changing those values
01:04.24ToAsTcfheh too?
01:08.48*** part/#htc-linux Spirits-Sight (~Spirits-S@c-24-91-187-210.hsd1.ma.comcast.net)
01:12.41mynwhat determines the div?
01:12.43*** join/#htc-linux marcin_ (~marcin@chello089078146089.chello.pl)
01:37.13ToAsTcfhbzo : i give up for tonight. maybe to fight it again tomarrow :P
01:38.38*** join/#htc-linux lancalot (~lancalot@74-92-235-219-York.hfc.comcastbusiness.net)
01:40.34*** join/#htc-linux MrPippy (~pip@adsl-75-42-163-160.dsl.sndg02.sbcglobal.net)
01:44.38*** join/#htc-linux Guest82173 (~compiler@the-compiler.org)
01:58.59*** join/#htc-linux The-Comp1ler (~compiler@unaffiliated/the-compiler)
02:01.11*** join/#htc-linux hceylan (~hceylan@83.66.117.84)
02:01.23*** join/#htc-linux The-Comp1ler (~compiler@the-compiler.org)
02:01.23*** join/#htc-linux The-Comp1ler (~compiler@unaffiliated/the-compiler)
02:04.03*** join/#htc-linux krisje8 (krisje8@gateway/shell/blinkenshell.org/x-ksjnvbijckspzazi)
02:17.51*** join/#htc-linux davep (~ohno_@ppp121-45-178-149.lns20.syd7.internode.on.net)
02:32.01*** join/#htc-linux 40FAAH91G (~tmzt@adsl-99-164-34-42.dsl.akrnoh.sbcglobal.net)
02:53.32*** join/#htc-linux tmzt_ (~tmzt@adsl-99-164-34-42.dsl.akrnoh.sbcglobal.net)
03:00.10*** join/#htc-linux OpnSrc1 (~harisha1@c-24-5-9-136.hsd1.ca.comcast.net)
03:07.33*** join/#htc-linux jumoit (~samuel@121.77.87.145)
03:14.26*** join/#htc-linux Unholy (~IceChat7@24.227.27.77)
03:26.57*** join/#htc-linux Zoolooc_ (~fredsibar@p54957F9F.dip.t-dialin.net)
03:32.43*** join/#htc-linux babijoee (~babijoee4@d110-33-179-143.sun801.vic.optusnet.com.au)
03:34.01babijoeeMarkinus: ping
03:40.10stinebdbabijoee: fyi i'm mirroring http://htcandroid.xland.cz/ at http://xdandroid.southcape.org/
03:40.52babijoeestinebd: already updated on first post
03:40.53babijoee:)
03:40.59stinebdbabijoee: thanks
03:41.00babijoeei'm preparing new build
03:41.17babijoeejust need to get something from markinus
03:42.21stinebdok
03:42.32stinebdi tried to get launcher2 working but failed miserably :(
03:43.24babijoeelol
03:43.32babijoeewe'll need 2.1 full
03:43.42stinebdwon't work even with that
03:43.47stinebdi have a local build of 2.1
03:43.55babijoeestinebd: i'm going to add topaz gsensor to rootfs
03:43.59stinebdit doesn't support opengl es 1.0 lite, which is what our driver provides
03:44.00babijoeedunno how to add that to git
03:44.05babijoeeah k
03:44.32stinebdwhere is that gl driver from anyway? proprietary?
03:44.43babijoeeno clue
03:45.11babijoeeanyways i'm implementing it like the way we implemented custom kb layout
03:45.21stinebdhmm
03:45.33babijoeeas its a custom sensor file
03:45.33stinebdgsensor as in sensors.blahblah.so?
03:45.38babijoeeyeah
03:45.41stinebdok
03:45.46stinebdyou can do that for now but there's a better way
03:46.00babijoeebetter is to integrate both ?
03:46.27babijoeei'd prefer to do it the better way now :p
03:46.27stinebdwell the better way involves putting it in /system/lib/ and name it something like sensors.topaz.so
03:46.34babijoeeooo
03:46.36stinebdand then modifying something in build.prop or something
03:46.40stinebdi forget where exactly
03:46.44babijoeeah k
03:46.51babijoeei'll have a look where sensors are called from
03:47.01stinebdit might even be in init.eclair.rc
03:47.23stinebdand then you can change whatever that line is in our rootfs init/
03:47.33stinebderr /init... via sed
03:48.04stinebdthat's what we're going to have to end up doing with bluetooth
03:51.57babijoeebah cant find it :p
03:52.06babijoeei'll just do other method for now
03:52.24babijoeedoesn't seem to hard to add that line to startup.txt for topaz users
03:54.57stinebdah, HAL tells it what to look for (other than default)
03:55.06stinebdi'll have to work that out next week
03:55.13babijoeeah k
03:55.22babijoeeif you have time stinebd can you help me with git?
03:55.32stinebdyeah
03:55.36babijoeecool
03:55.37babijoeethanks
04:10.48*** join/#htc-linux Unholy (~IceChat7@24.227.27.77)
04:15.03babijoeestinebd: any awesome apps that you think that should be included in new release?
04:16.50stinebdbabijoee: not really, i'm a bit of a minimalist
04:17.01babijoeeah k
04:17.16babijoeei've just added a battery % meter in notification panel
04:17.27stinebdoh yeah
04:17.30babijoeegives actual percentage
04:17.32stinebdi have one from darshan on market
04:17.39stinebdbattery indicator lite or something
04:17.42babijoeeits called battery indcater lite
04:17.43babijoeeyep
04:17.47babijoeeawesome app
04:17.53stinebdyup
04:18.02babijoeei need things like those
04:18.07babijoeeanything else you use?
04:18.25babijoeei also see changes to camera in the init.cfg
04:18.26stinebdgenie widget
04:18.28babijoeeis camera working :p
04:18.33stinebdno camera yet
04:18.36babijoeegenie widgit
04:18.41babijoeehmmm i wonder if it will work
04:18.48stinebdgenie widget is the news & weather thing
04:18.53babijoeeyeah i know
04:19.14stinebdthat, battery indicator and home++ are the only things i usually have to install after clearing data
04:19.46stinebdoh, and k9 mail for push imap
04:20.30babijoeehome++ isnt free?
04:20.46babijoeeis*
04:20.52stinebdthe beta is
04:21.32babijoeehrmmm
04:21.36babijoeewats the limitations?
04:21.46babijoeeany anoying time limits?
04:22.09stinebdnot that i know of
04:22.12babijoeeah k
04:22.17babijoeei'll check it out
04:22.46*** join/#htc-linux vicbush (~45422a91@gateway/web/freenode/x-wcdcpsgjwgbbfhyr)
04:25.54babijoeealso to double check
04:25.59babijoeedoes this code look right
04:26.00babijoeeif /bin/grep -c 'sensor=topaz' /proc/cmdline >/dev/null ; then
04:26.00babijoeeecho "USING topaz sensor"
04:26.01babijoeecp -f /lib/hw/sensors.topaz.so /lib/hw/sensors.default.so
04:27.23tmztln -s ?
04:28.18babijoeeanyone with topaz around?
04:28.35stinebdyou could use ln -sf, yes. also don't forget the 'fi' after all that
04:29.34babijoeeln -s/f where?
04:30.02stinebdinstead of cp -f
04:30.08babijoeeooo
04:30.29babijoeetmzt didnt know you were talking to me.. but then again no1 is online here :P
04:30.36babijoeeah k
04:31.09babijoeewill it still work with cp -f?
04:31.20babijoeeor is ln -sf better?
04:31.30stinebdit'll work, but take up more space
04:31.37babijoeeah k
04:31.43babijoeeso ln -sf better
04:31.55stinebdyeah
04:33.19babijoeethanks
04:34.49babijoeestinebd: would you say home++ is smoother?
04:35.00stinebdslightly
04:35.13babijoeei'm thinking of whether to use it as main or put in androidapps folder
04:35.29stinebdi'd say put it in androidapps
04:35.33*** join/#htc-linux Neo31 (~Neo31@unaffiliated/neo31)
04:35.42babijoeewoah 3-7 screens
04:35.42babijoeenice
04:35.53babijoeei might use it as main :p
04:35.58babijoeelets see what people think
04:36.05babijoeei'm getting bored of old look
04:38.39babijoeeit loops the navigation panels :P
04:40.05babijoeeyeah i'll leave it androidapps
04:41.41babijoeenope i'm using it as main
04:41.44babijoeeits pretty awesome
04:52.15Unholyhome++ crashes alot
04:52.22Unholythats y i stopped using it
05:08.38*** join/#htc-linux babijoee (~babijoee4@202.124.88.68)
05:16.47AstainHellbringhi cr2
05:16.52AstainHellbringhi all
05:18.35*** join/#htc-linux fleixius_ (~fleixius@c-68-55-69-186.hsd1.md.comcast.net)
05:18.51stinebdi gotta hit the sack
05:20.28*** join/#htc-linux babijoee2 (~babijoee4@203.171.196.158)
05:23.10babijoee2bb stinebd
05:28.07*** join/#htc-linux jamenlang8 (~jamenlang@68.170.60.12)
05:36.19randomblameyay mddi cmd send rtd: int 23a000, stat 808063, rtd val e
05:36.30randomblameone message closer to it working
05:36.45randomblamethats the same values as the topaz, I wonder if that's significant
05:50.29tmztwhy are you still initing it?
05:50.41tmztwhere arfe you getting this output?
05:50.49tmztwhen you mess with mddi htcfbcon dies
05:58.25*** join/#htc-linux Unholy (~IceChat7@24.227.27.77)
06:10.04randomblamezomg
06:10.25randomblameI got mddi on and a flickering white and grey pannel
06:10.31randomblameprogresss
06:10.59tmztgood
06:11.08tmztbut why are you initing the lcd at all?
06:11.24tmztif you don't power it off you shouldn't have to
06:11.32tmztis the resolution correct?
06:14.40randomblamepsh theres nothing on screen i may have been fooled
06:14.57randomblameit's nothing readable or recognizable
06:15.09tmztno flickering is good
06:15.43randomblameramlog doesn't say what I wanted it to though
06:15.45randomblamemddi: init() base=0xc500e000 irq=16
06:15.45randomblame[   35.884643] smd_alloc_channel() 'SMD_RPCCALL' cid=2
06:15.45randomblame[   35.945220] mddi cmd send rtd: int 23a000, stat 808063, rtd val e
06:15.45randomblame[   36.005584] RPC_dev init complete!
06:15.45randomblame[   36.065307] Unable to handle kernel NULL pointer dereference at virtual address 00000000
06:16.09ToAsTcfh:/
06:16.23randomblamemeaning it didn't get further than last time, rpc stuff was going on at same time
06:17.51randomblametmzt I'm trying to do it with panel drivers and all this crap because without it msmfb never starts
06:17.54tmztyep still panic
06:18.18tmztdid you #if 0 the panel drivers?
06:18.27randomblameI didn't understand that
06:18.30tmztdid you prevengt it from powering down the vreg
06:18.38tmztall the init code
06:18.46tmztwhere it send stuff to the panel
06:18.52tmztput #if 0
06:18.55tmzt#endif
06:18.57tmztaround it
06:19.05tmztincluding the vreg call
06:19.14randomblameah, I made my own board panel with the minimums
06:19.47randomblamemeaning no power client
06:20.01randomblameand hopefully that means mddi.c wouldn't attempt to power it down
06:20.56randomblamebrb
06:26.08randomblamewouldn't the panel need to be reinitialized? because mddi.c and msm_fb get their memory addresses from pmem.c which is not the address that wince has the frame buffer on
06:26.20randomblameso it would have to be restarted on that memory address to work correct?
06:28.38*** join/#htc-linux MrPippy (~pip@adsl-75-42-163-160.dsl.sndg02.sbcglobal.net)
06:30.08randomblamespeaking of the devil, mr pippy if I wanted to start using my lcd panel without reinitializing it, because I have no panel driver wouldn't the frame buffer need to continue writing to the correct memory address
06:30.16*** join/#htc-linux Reefermattness (~MattGNM@c-76-18-79-170.hsd1.nm.comcast.net)
06:44.17randomblameI guess I need to go through and // some of mddi.c so it will freaking continue
06:59.11*** join/#htc-linux nacho-g (~nacho-g@190.178.178.248)
07:11.01*** join/#htc-linux babijoee (~babijoee4@203.171.196.72)
07:11.28*** join/#htc-linux GNUtoo (~GNUtoo@host192-8-dynamic.9-87-r.retail.telecomitalia.it)
07:17.18*** join/#htc-linux jamenlang8 (~jamenlang@68.170.60.11)
07:18.06randomblameok there we go
07:18.08randomblame[    7.774230] mddi: registering panel mddi_c_dummy
07:18.08randomblame[    7.803161] mddi: publish:
07:18.08randomblame[    7.844238] msm_serial: driver initialized
07:19.36randomblameand it continues until later crash but never switches consoles just grey flickering screen
07:19.51randomblamemddi_c_dummy is broken right?
07:29.24randomblamewhat starts msm_fb.c? it never starts
07:30.29randomblameah it starts itself
07:43.32*** join/#htc-linux babijoee (~babijoee4@d110-33-179-143.sun801.vic.optusnet.com.au)
07:45.04randomblamemsmfb_probe: no fb_data
07:48.04*** join/#htc-linux droid001 (~g1@p4FDC9BEB.dip.t-dialin.net)
07:48.19*** join/#htc-linux fleixius_ (~fleixius@c-68-55-69-186.hsd1.md.comcast.net)
07:48.29*** join/#htc-linux fleixius (~fleixius@unaffiliated/fleixius)
07:56.39*** join/#htc-linux balans (~balans@82-170-215-5.ip.telfort.nl)
07:59.41*** join/#htc-linux balans (~balans@82-170-215-5.ip.telfort.nl)
08:11.03*** join/#htc-linux Wout|afk (~5070d0f6@gateway/web/freenode/x-ihufoxsilhknupto)
08:17.29*** join/#htc-linux dream_kill (~nospam@server88-208-237-236.live-servers.net)
08:23.17*** join/#htc-linux kiozen (~oeichler@rgnb-5d87d14c.pool.mediaWays.net)
08:29.58*** join/#htc-linux M1DLGpc_ (~M1DLGpc@bb-87-81-252-83.ukonline.co.uk)
08:40.29*** join/#htc-linux M1DLGpc_ (~M1DLGpc@bb-87-81-252-83.ukonline.co.uk)
08:44.17*** join/#htc-linux SOG (~SOG@119.237.154.191)
08:45.57randomblamenow it hangs after installing 240x400 panel, must be the dummy client
08:46.15*** join/#htc-linux bzo (~chatzilla@adsl-75-35-5-163.dsl.pltn13.sbcglobal.net)
08:49.59*** join/#htc-linux mickey|bbl (~mickey@openmoko/coreteam/mickey)
09:00.40*** join/#htc-linux M1DLGpc_ (~M1DLGpc@bb-87-81-252-83.ukonline.co.uk)
09:14.38*** join/#htc-linux sxe (~quassel@ip-62-143-102-161.unitymediagroup.de)
09:17.43*** join/#htc-linux Markinus (~Miranda@gtng-4db048e3.pool.mediaWays.net)
09:53.22*** part/#htc-linux jumoit (~samuel@121.77.87.145)
09:57.17*** join/#htc-linux Captnoord (~Captnoord@dc5147a47b.adsl.wanadoo.nl)
10:02.09*** join/#htc-linux kvaster (~kvaster@live.bn.by)
10:06.46MrPippyphh: that xda post mentioning cdma data on rhod is just talking about raph, his sig even says "touch pro"
10:07.51Captnoordlol
10:08.10tmztyeah
10:08.18tmztsadly
10:08.32tmztthough ril should work the same on rhod
10:18.47MrPippyhave you discovered anything more about why it doesn't work?
10:20.28tmztme?
10:20.44tmztit just wants there to be a sim then registers with the gsm network
10:21.03tmztI don't have the traces for at to go into or out of global mode
10:21.53MrPippydon't have a sim to test with?
10:22.11tmztyes but only voda
10:22.21tmztvoice calls work though
10:22.25tmztwhich is weird
10:22.36MrPippygoing through gsm or cdma?
10:22.41tmztso it's probably just trying to creat a gprs context
10:22.43tmztcdma
10:23.31MrPippyyeah weird
10:27.33MrPippyi'm really really close to getting wifi working (including android changes), once thats done i'll look at the cdma data
10:29.44*** join/#htc-linux GeertJohan (~Squarc@82-217-32-29.cable.quicknet.nl)
10:33.55MarkinusMrPippy: you're on WIFI for Rhod? Did you have the problem with crashing on disabling of WIFI too?
10:34.42MrPippyi had that problem on diam (and found a possible fix), rhod is broadcom wifi
10:36.38MarkinusMrPippy: I great. . . I fix would be good ..
10:36.39MarkinusA
10:39.04MrPippytry reverting this merge http://android.git.kernel.org/?p=platform/system/wlan/ti.git;a=commitdiff;h=849c6d4186a7fcd4d25bcff6cc2e6bc25f8c4696
10:39.26MrPippyso get rid of the release_host and claim_host calls
10:40.35MarkinusMrPippy: Ok, I will try. In the Moment we have to deactivate the disabling of vregs, then this works on TOpaz. But it could be a random, think . .
10:41.32MrPippyohh yeah i think i had that problem too on rhod
10:41.45MrPippyturn off the vregs and it just resets, arm9 must not be happy with that
10:42.55MarkinusMrPippy: yes, but I'M sure that this worked on first Topaz releases . .
10:43.18MarkinusMrPippy: something is changed but I'Ve no Idea what+
10:44.48Captnoordlol reboot
10:44.50Captnoordloooool
10:44.53Captnoordreboot into android
10:44.54Captnoord:S
10:46.26*** join/#htc-linux skodde (~skodde@unaffiliated/skodde)
11:00.05*** join/#htc-linux phh (~quassel@4be54-3-82-228-187-43.fbx.proxad.net)
11:02.16MrPippyNICE got wifi working on rhod in android
11:02.47phhcool
11:04.23phh04:43babijoeestinebd: i'm going to add topaz gsensor to rootfs <-------- ?!?
11:04.39phhwhy in *rootfs* ? the driver is in system
11:04.42phhand I already added it
11:05.17randomblameI'm glad you guys are working so hard, by the time I get my frame buffer working I'll realize that 240x400 isn't enough for android and I'll have to buy an htc phone
11:05.33phhlol
11:05.51phhrandomblame: kaiser guys use scaling
11:05.55phhyou can do the same
11:06.03randomblameone day
11:06.44randomblameif the dummy client is loaded I hang at msm fb, if any other client is loaded I never see msmfb init
11:06.51phh04:45babijoeeas its a custom sensor file <------- lol ?
11:06.58phhI did my best so that it's the same for everyone
11:08.41*** join/#htc-linux dream_kill (~nospam@server88-208-237-236.live-servers.net)
11:08.58phhwhat the hell has he done ?!?
11:09.00MrPippyweird the market doesn't work, it just throws a server error
11:09.59*** join/#htc-linux ccube (~ccube@95-90-193-182-dynip.superkabel.de)
11:11.53*** join/#htc-linux BabelO (~fcr@2a01:e35:2ee1:c1c0:21e:8cff:fe2d:7d3f)
11:11.54*** join/#htc-linux BabelO (~fcr@unaffiliated/babelo)
11:15.18MrPippyphh: i think these lines can be pulled out of /init
11:15.22MrPippy#Activate wifi
11:15.23MrPippyecho 1 > /dbgfs/htcraphaelmmc_dbg/wifi_pwr
11:15.23MrPippyecho 1 > /dbgfs/htcraphaelmmc_dbg/wifi_cd
11:15.29phhMrPippy: yup.
11:16.50Captnoordhmmm wifi off stalls
11:18.48phhMrPippy: about market, does your ril have any way to auth you ? I think the market uses imei or something like that
11:19.52MrPippynah the phone isn't activated, i don't think the ril is working at all
11:20.25MrPippythe logcat errors weren't helpful at all com.android.vending.api.FatalCommunicationException: Sent a bad request.
11:20.56phhok
11:21.25*** join/#htc-linux wout (~5070d0e2@gateway/web/freenode/x-rhuodbzdnokjuldy)
11:22.14MrPippystinebd: can i get commit access for rootfs? i'll have a patch for rhod wifi ready in a few hours
11:27.11Markinusphh: something in the last commits kills WLAN on Topaz
11:27.23Markinusphh: it'S reseting on start of WLAN
11:27.37Captnoordsame goes for me... I can't disable wlan
11:27.42CaptnoordI hope that it isn't my patch
11:27.44Captnoord:P
11:28.08Markinushmm, I had never problems with starting . . .
11:34.33CaptnoordI think it are the net: wireless: bcm4329: commits
11:34.36Captnoordobviouse
11:34.36Captnoord:P
11:34.53CaptnoordI wonder in what way the userland modules are handling this
11:35.11Captnoordif the kernel and module are mixed up
11:35.17Captnoordwe can expect this
11:39.32*** join/#htc-linux M1DLGpc__ (~M1DLGpc@bb-87-81-252-83.ukonline.co.uk)
11:42.31MarkinusCaptnoord: the  bcm4329 commits do nothing for out TI wlan . . hmm..  this has to be somethink else
11:43.00Captnoordk
11:43.01*** join/#htc-linux toi (~toi@d54C2A96D.access.telenet.be)
11:43.08Captnoordcan you remember which commit worked
11:43.09Captnoord?
11:43.58MarkinusCaptnoord: nop, maybe we can see it on glemson . ..
11:45.18MarkinusCaptnoord: intresting . .the last Build seems to work on a other TOpaz . .. hmm
11:45.59phhhaha.
11:46.33Markinusthis is one of this very strange things . . .
11:46.48Captnoordnever trust users
11:46.56Markinushmm, yes .. right
11:47.05Markinus:P
11:47.38phhMarkinus: are you aware of babijoee's changes for sensors ?
11:48.05phhit seems he made a script to detect which driver to use.... which is totally stupid since our driver works for everyone
11:48.27Markinusphh: urgh ..  .
11:48.45phhI haven't seen the commit yet, some little problems on my desktopo
11:48.50Markinusphh: This Ideas wasn'T from me :)
11:50.14Markinusphh: I send him my binary sensor file, maybe he thought we need both. . .
11:50.27*** join/#htc-linux phh (~quassel@2a01:e35:2e4b:b2b0:250:8dff:fee1:c793)
11:50.46Markinusphh: I send him my binary sensor file, maybe he thought we need both. . .
11:51.24phhreading the logs, it seems so
11:58.28randomblamehmm using mddi_client_nt35399 I get some interesting messages
11:58.28*** join/#htc-linux sektorNBA (ricardo@200.143.33.138)
11:58.28*** join/#htc-linux sektorNBA (ricardo@unaffiliated/sektornba)
11:58.36randomblamevsync on gpio 97 now 0
11:59.01randomblameobviously it misdetects the gpio it should be, but it knows the gpio it was set to which is the correct one
11:59.06phh?
11:59.10phhyour vsync gpio is != 97 ?
11:59.14phhit's hardcoded in the client file
12:00.02randomblamewhy's everything gotta suck
12:00.17randomblamemy service manual doesn't tell me the vsync gpio
12:00.48randomblameand I don't know how the hell to go about getting it out of haret
12:01.11phhassume it's 97.
12:01.22phhit's so for all boards i've seen
12:01.30randomblameyou always say the nicest things, I love that idea
12:02.03randomblameso I have to hard code it into the client
12:02.07phhit's already
12:02.14randomblamethen why does it say now 0
12:02.26phhin haret, watch gpios, and see if there is a gpio that runs at ~ 60Hz
12:02.31phhbecause it can be set to 1 or to 0
12:02.33phhie up or down
12:02.47randomblamelol
12:02.53randomblamekk
12:03.14randomblameit's 4 am I was hoping to be able to glean information from the output,
12:03.32randomblamenaive of me
12:08.30phhMarkinus: he didn't even commit to the rootfs thing ...
12:10.53*** join/#htc-linux droid (~droid@061093132071.ctinets.com)
12:11.00Markinusphh: hmmm
12:11.28Markinusphh: I see in the 27.02 is something about Bluetooth, is BT now working on 2,01?
12:11.44phhyeah stinedb got it working
12:11.52Markinusphh: ahh, cool!
12:11.55phhno A2DP though, I killed it when I got sound working.
12:12.05phhit needs to rebuild the audio libs with the appropriate flag
12:13.04randomblamephh, grey screen just before msm_fb initiates normal?
12:13.07Markinusphh: ah, ok, if we know what we have to do . . . .:) all is good!
12:13.11phhrandomblame: don't know
12:13.16phhMarkinus: no.
12:13.40phhwell I think stinebd should be able to get it working as he already built an android
12:13.46phhbut I haven't ever been able to :/
12:13.50randomblamewas hoping htc devices had the same behavior with mddi on and no frame buffer writing to it
12:14.00randomblameit would give me hope
12:14.29Markinusphh: ohh
12:17.04Markinusphh': is this the libaudio from android?
12:17.33phhit's libaudiosomething
12:17.38phhlibaudioflinger I think
12:21.33*** join/#htc-linux BabelO (~fcr@unaffiliated/babelo)
12:25.49Markinusphh: I look to the liblary and there are the flags for AD2P in . . . do you know the correct flag that we needed?
12:26.02phhno :(
12:26.08phhwell look at Android.mk
12:26.21phhthe flag is there, but then you'll need to build bluez and many other deps
12:26.40randomblame<PROTECTED>
12:26.43randomblamenight everybody
12:26.52phhit's 1pm ! :p
12:27.02randomblameno it's 4:27
12:27.03Markinusphh: thgis is in:   LOCAL_CFLAGS += -DWITH_BLUETOOTH -DWITH_A2DP
12:27.35randomblame4:27 am ha!
12:27.41phhMarkinus: there is a if somewhere no ?
12:28.00Markinusq ($(BOARD_HAVE_BLUETOOTH),true)
12:28.13phhso you have to set this to true (or comment the if ? :D)
12:28.27Markinusphh: if this is alll :)
12:28.47phhbut check stinebd's gitorious, I think he has a big mm command which should help
12:30.52phh#Topaz sensor
12:30.52phhif /bin/grep -c 'sensor=topaz' /proc/cmdline >/dev/null ; then
12:30.52phh<PROTECTED>
12:30.52phh<PROTECTED>
12:30.53phhfi
12:30.55phhtss
12:30.59phhnot even proper detection
12:34.38*** join/#htc-linux Reefermattness (~MattGNM@c-76-18-79-170.hsd1.nm.comcast.net)
12:34.53Markinusphh: ohh, over command line. .
12:35.00phhyes ...
12:35.08phhwhile we already have mtype detection in the script
12:35.22phhpff he is maybe not a bad packager, but he'd better stop here -_-'
12:39.36*** join/#htc-linux SOG (~SOG@119.237.154.191)
12:48.03phhcr2 extracting boot.img with some od -x and comparing with a gzip header is easy
12:51.21*** join/#htc-linux Phil_123 (~anonymous@221.127.43.91)
12:54.08Phil_123any devs (hi dzo!) working on porting the kaiser camera (s3k3c1fx) to eclair?
12:55.09Phil_123i stumbled upon a phone camera testing device which supports many cameras, including s3k3c1fx
12:56.08Phil_123will it be easier to disassemble its driver and see how it interfaces with s3k3c1fx and put it into the linux driver?
12:57.31*** join/#htc-linux zachy (~george@85.193.4.18)
12:59.15Captnoorddisassemble is never easy
13:00.17phhPhil_123: why disassemble ?
13:00.21phhAFAIK all sources are available
13:01.18Phil_123the driver for that phone camera testing device is windows only
13:01.47phhthere are open drivers (maybe not s3k3c1fx, but friends) for linux
13:02.21Phil_123i see
13:03.48Phil_123my kaiser runs wm6.5 and it sucks compared to wp7, so i plan to flash android to it but the 2.1 builds don't have the camera working
13:04.19phhI know
13:04.22phhnoone has camera working on 2.1
13:04.26phhI mean no msm7k
13:04.26Phil_123and from what i read (but not tried yet) the 1.6 builds don't work as well as wm does
13:09.01*** join/#htc-linux cr2_ (~cr2@ip-109-84-155-132.web.vodafone.de)
13:09.44cr2_good morning :)
13:09.48Phil_123i'm no novice when it comes to programming software in java, javascript or php, but i'm a total noob when it comes to programming code for hardware devices
13:10.26Markinuscr2_: good morning! This was a long night he? :)
13:10.29*** join/#htc-linux babijoee (~babijoee4@d110-33-179-143.sun801.vic.optusnet.com.au)
13:10.35babijoeehey guyd
13:10.44babijoeeguys*
13:10.45Phil_123i use linux sometimes and understand how difficult things can be to get hardware working in linux when the datasheets aren't very useful and there's no help from the manufacturer
13:11.13cr2_NetRipper: all *MDDI* code should go away from your console driver
13:11.46cr2_Markinus: yeah :) and the weather is nice, will walk around a bit later
13:11.58cr2_babijoee: do you have nexus ?
13:12.05babijoeecr2 yes'
13:12.17cr2_babijoee: and devmem2 ? :)
13:12.27babijoeedevmem2?
13:12.28babijoeedunno
13:12.31babijoeewats that
13:12.45Phil_123where and how can i start learning how can i get hardware to work in android? specifically, getting the camera to work in android 2.1? what problems are we having with it
13:13.01cr2_babijoee: a linux program to read from /dev/mem
13:13.06Markinuscr2_: Yes, we did It already . . very nice. .. so without snow :)
13:13.07babijoeei'll dl it
13:13.11Phil_123*kaiser's s3k3c1fx camera
13:13.18CaptnoordPhil_123: you know how to code?
13:13.25Captnoordjust start reading the source code
13:13.27Captnoordstick around
13:13.31Captnoordand google a lot
13:13.31cr2_Markinus: let's fix the dma setup
13:13.34Phil_123Captnoord: yes but not C
13:13.38Captnoordthen
13:13.39Markinuscr2_:  ok
13:13.43Captnoordfirst learn C
13:13.45Captnoordsmall things first
13:13.48Captnoordvery important
13:14.04Captnoordnever attempt something big before you know anough about a coding language
13:14.15babijoeeMarkinus: have you tested new build>
13:14.16babijoee?
13:14.19cr2_Markinus: i've decoded the bit manually, and got differences with nexus
13:14.41babijoeei've added the sensors for topaz
13:14.50Markinusbabijoee: yes, but you noit understand me right :) We can user this sensor file for all devices . .
13:14.51cr2_Markinus: need some independent verification
13:15.03Markinusbabijoee: it'S the same with small modifications for Topaz
13:15.04babijoeeMarkinus
13:15.08babijoee-.-"
13:15.08Phil_123i learnt C before, moved onto other languages, so i'm not good at things C coders have to handle, such as pointers and memory allocation and stuff
13:15.16babijoeei didnt know that :p
13:15.26babijoeei'll update it next build :0
13:15.28Phil_123and i have never coded drivers before
13:15.41CaptnoordPhil_123: if you want to help improving support
13:15.43Captnoordyou need to learn
13:15.55Captnoorddude i've never coded drivers before
13:16.03Phil_123would looking at source code of drivers for existing devices help?
13:16.03Captnoordbut that doesn't make me unable todo so
13:16.06Markinuscr2_: ok
13:16.10Captnoordit helps as a reference
13:16.20Captnoordbut unless you understand fully whats going on
13:16.26Captnoordits not really much use
13:16.27Captnoordso
13:16.27Phil_123i have some C code i coded many years ago
13:16.29Captnoordgo learn c
13:16.33Captnoorduse c
13:16.43CaptnoordI learned c/cpp by coding everyday
13:16.45Captnoordsmall apps
13:16.46Captnoordsmall things
13:16.49cr2_Markinus: 02102196
13:16.50Captnoordand later moved to bigger things
13:17.18cr2_Markinus: 0000.0010.0001.0000.0010.0001.1001.0110
13:18.13cr2_Markinus: bits 1,2,4,7,8,13,20,25
13:18.54babijoeecr2: i'll have to go out for abit. I'll should be back in an hour
13:18.56cr2_Markinus: drivers/video/msm/mdp_hw.h
13:19.02babijoeeis this nexus one stuff urgent?
13:19.09Markinuscr2_: sry, I'M in the moment not by you ?!? bits from what?
13:19.20phhCaptnoord: can you disassemble htc_acoustic and see it writes which file's infos to which memory area ?
13:19.21Markinuscr2_: ahh, ok
13:19.27cr2_babijoee: not urgent, but very helpful
13:19.40Captnoordi'll check
13:19.40cr2_Markinus: MDP_DMA_P_CONFIG
13:19.50babijoeecr2: okay i'll definantly help out in any way i can
13:19.54babijoeebbl guys
13:20.22Phil_123so assume that i got my C knowledge back now, before attempting to work on s3k3c1fx drivers, looking at drivers which use the i2c protocol is a good place to start right?
13:21.54phhCaptnoord: it does a mmap to /dev/htc_acoustic, then writes to its memory, then do an ioctl
13:22.44Phil_123thank you guys for the advice
13:22.52cr2_Markinus: 25 is #define DMA_IBUF_FORMAT_RGB565                  (1 << 25)
13:23.08phhPhil_123: i'd say start reading your current s3k3 thing driver
13:23.22cr2_Markinus: 20 is #define DMA_OUT_SEL_LCDC                        (2 << 19)
13:23.48Captnoordhtc_acoustic.so?
13:24.21Captnoordor wince dll?
13:24.21phhCaptnoord: yes
13:24.23Captnoordk
13:24.23phh.so
13:24.54cr2_Markinus: 1 is #define DMA_DSTC0G_6BITS (2<<0)
13:25.05Markinuscr2_: yes yes I'm on it . . :) I reading the code ..
13:25.23cr2_Markinus: 2 is #define DMA_DSTC1B_5BITS (1<<2)
13:26.05cr2_Markinus: 4 is #define DMA_DSTC2R_5BITS (1<<4)
13:26.35Captnoordphh send me yours
13:26.40cr2_Markinus: 7 is #define DMA_PACK_ALIGN_MSB (1<<7)
13:26.44Captnoordotherwise we don't have the same resource file
13:26.48cr2_Markinus: nexus is LSB
13:27.15phhCaptnoord: http://husson.hd.free.fr/libhtc_acoustic.so
13:27.16cr2_Markinus: 8 and 13 now :)
13:28.03Markinus#define DMA_PACK_PATTERN_MASK (0x3f<<8)
13:28.26*** join/#htc-linux wout (~5070d0e2@gateway/web/freenode/x-nxknwqxhawjfgwzd)
13:28.36cr2_yes, to clean the whole range
13:29.38Markinuscr2_: and the to set the RGB and BGR
13:29.42cr2_Markinus: sorry, it was not nexus, but the code of NetRipper. taken from g1 probably
13:29.57cr2_Markinus: yes, there are 2 options
13:30.09cr2_the current ode has DMA_PACK_PATTERN_RGB
13:31.00cr2_need to convert it into bits
13:31.35Captnoordphh check
13:31.35Captnoordsub_2334
13:32.05CaptnoordI guess that memset 0
13:32.16Captnoordall the comming mem regions
13:32.38Captnoordi'm talking in
13:32.38Captnoordset_acoustic_parameters
13:32.44Captnoordabout
13:32.48cr2_Markinus: nexus sets DMA_PACK_ALIGN_LSB, DMA_PACK_PATTERN_RGB, DMA_DITHER_EN
13:33.00Captnoordthen last jump addr loc_2F76
13:33.04Captnoordis where the magic happens
13:33.16Captnoordthat function should be called: android__check_and_set_audpp_parameters
13:33.18Captnoordor
13:33.23Captnoordandroid::check_and_set_audpp_parameters
13:34.03Captnoordand where you see
13:34.04CaptnoordBL      _ZN7android30check_and_set_audpp_parametersEPci ; android::check_and_set_audpp_parameters(char *,int)
13:34.08Markinuscr2_: ok I do it
13:34.12Captnoordis where the mem copy is located
13:34.32cr2_Markinus: can try together
13:34.51cr2_Markinus: RGB is
13:34.59GNUtoohi leviathan
13:34.59Captnoord./dev/msm_pcm_ctl
13:35.11MarkinusDMA_DSTC0G_6BITS -> is set
13:35.35MarkinusRGB:
13:35.40GNUtooleviathan, what do you think: what would be quicker: remove pmem from .31 msmfb or port a lot of things to .32 msmfb
13:35.46Markinus0.0001.1
13:36.05cr2_Markinus: 2<<(2*2)|0<<2|1
13:36.41cr2_5 and 0 (+8)
13:36.43cr2_8 and 13
13:36.51cr2_so we have RGB
13:37.11Markinuscr2_: I need a moment to understad :)
13:37.20cr2_Markinus: so nexus sets DMA_PACK_ALIGN_LSB, DMA_PACK_PATTERN_RGB, DMA_DITHER_EN
13:37.52cr2_Markinus: we have DMA_PACK_ALIGN_MSB, DMA_PACK_PATTERN_RGB and no DMA_DITHER_EN
13:38.25cr2_weird. so we need to flip R and B
13:38.40cr2_otherwise get pseudocolors :)
13:38.45Markinus:)
13:38.59Markinusbecaus LSB/MSB yes?
13:39.01cr2_err
13:39.13cr2_yes, because of MSB
13:39.22*** part/#htc-linux zachy (~george@85.193.4.18)
13:39.40cr2_Markinus: but it's nt important for the console
13:39.48leviathanGNUtoo: I would like to import the missing functions into the alread working msmfb driver of our branch
13:39.49cr2_Markinus: which is b/w
13:40.10GNUtooleviathan, that would make us a lot of work no?
13:41.02Markinuscr2_: But if we would have A frambuffer logo ..
13:41.12cr2_Markinus: androids again put per-panel parameters as generic ;)
13:41.31cr2_Markinus: yeah, then it will be wrong
13:41.58cr2_Markinus: the blac setup had such problems
13:42.37Markinuscr2_: maybe the same.  .
13:42.45cr2_mdp_writel(NULL, 0, MDP_DMA_P_START);
13:42.55cr2_this is set by NetRipper
13:43.02cr2_need to think about it
13:43.04Markinuswe need 1 . . .
13:43.17cr2_if we need it at all
13:43.18Markinusor?
13:43.24GNUtooleviathan, mfd which is a variable from a struct that we don't have seem in all the .31 driver
13:43.35cr2_need to check uboot code
13:43.54cr2_Markinus: on blac it was also 565 vs 666 setup
13:44.14NetRippererrr there's no need to modify htc_fb_console
13:44.27NetRipperor werent you talking about that console driver cr2?
13:44.58cr2_NetRipper: about htc_fb_console, which is not htc_fb_console
13:45.22cr2_NetRipper: have you removed all mddi references from it ?
13:45.23GNUtooleviathan, I am also afraid that we would have to import most of the driver,are you shure the issue is only because of refresh
13:45.24GNUtoo?
13:45.34NetRippercr2_, what mddi references?
13:46.18NetRippercr2_, the mddi params that are written?
13:46.22cr2_<PROTECTED>
13:46.22NetRipper3 of them i think
13:46.23cr2_<PROTECTED>
13:46.25cr2_<PROTECTED>
13:46.30NetRipperwe can try removing them yes
13:46.49cr2_NetRipper: and you can do correct DMA setup too
13:46.54NetRipperwe only need the fbram, size, stride and do the dma
13:47.08NetRippercr2_, no, correct DMA setup is different per device/panel
13:47.12GNUtooleviathan, also we don't have the dma functions
13:47.27NetRippercr2_, and i want to make htc_fb_console generic, so that it can be used together with the VIDEO tag
13:47.32cr2_fbram code is too obfuscated now too. just pick the addr from VIDEOLFB :)
13:47.56NetRippercr2_, yes but that means we can't add dma config, as with a bad config it wouldnt work at all
13:47.59cr2_NetRipper: how will you do generic MDDI vs LCDC ?
13:48.02NetRipperor have bad colors etc
13:48.19cr2_NetRipper: then do switch(mtype)
13:48.19NetRippercr2_, just depend on wince/spl init
13:48.44cr2_NetRipper: you talk like a commercial developer on a deadline :)
13:48.53GNUtooleviathan, ping?
13:49.11NetRippercr2_, no the htc_fb_console is a hack, it doesn't need to be beautifl... msm_fb is the one we need to get working
13:49.37cr2_NetRipper: bt at least add the comments. because we already know how to do the right thing
13:49.50cr2_NetRipper: should i add lcd param decoder to haret ?
13:49.51GNUtooNetRipper, hi what are you doing with msm_fb.c ? for which device ? under which kernel?
13:50.01NetRipperGNUtoo, leo
13:50.09NetRipperGNUtoo, and im not doing anything to it atm..
13:50.09GNUtoook
13:50.15GNUtooah ok
13:50.29NetRippercr2_, there are far more pressing issues than the htc_fb_console
13:50.35NetRipperlike the irq problem
13:50.51GNUtoobecause we try to make xf86-video-msm work flawlessly
13:51.16NetRipperGNUtoo, we'd first need a proper kernel
13:51.24GNUtooah ok
13:51.29NetRipperi suspect xf86-video-msm relies on msm_fb
13:51.33GNUtooindeed
13:51.47NetRippercurrent MDP is not working, which is a prerequirement for msm_fb
13:51.59GNUtooMDP is?
13:52.11cr2_msm_fb ? isn't xf86-video-msm accelerated ?
13:52.14NetRipperthe thing that dma's buffers to the display
13:52.26GNUtoook
13:52.31cr2_GNUtoo: video data processor
13:52.36GNUtoothanks a lot
13:52.49cr2_GNUtoo: does rotation, scaling and alpha blending
13:52.51GNUtoocr2_, it isn't working for leo or for all
13:52.57GNUtoowow
13:53.07NetRipperGNUtoo, for leo.. it's working on msm7k
13:53.11GNUtoook nice
13:53.13NetRippernot on qsd
13:53.13GNUtoothanks a lot
13:53.15cr2_GNUtoo: we have some irq setup problem on leo
13:53.21GNUtoook
13:53.41NetRipperanyway im off till tomorrow night so i wont do much this weekend :)
13:53.49NetRipperbut i dont think you should concentrate on the htc_fb_console
13:53.53GNUtoowhat should I do then? import the working driver which is compatible with xf86-video-msm or patch the one we've got
13:53.53cr2_NetRipper: ok, let's thing how this 0xffff value appears in the irq setup
13:53.54GNUtoo?
13:54.53NetRipperGNUtoo, what device are you working on?
13:55.07GNUtoodream
13:55.16NetRipperthen you should ignore what cr2 and me say about the leo :)
13:55.18cr2_NetRipper: qsd sets some (smsm) irqs to 0xff, but this should be trapped by mach-msm/irq.c
13:55.21GNUtoook
13:55.28cr2_irq 65535
13:55.50GNUtoobut we've got gtk issues with xf86-video-msm with the current driver
13:55.59NetRippercr2_, maybe there's an error or maybe somewhere a signed short is used instead of unsigned
13:56.01GNUtoolike corupt display and similar
13:56.09NetRipperor signed int instead of unsigned short i mean
13:56.10cr2_GNUtoo: we can't even properly boot with a ramdisk now
13:56.11GNUtoonot with enlightenment applications tough
13:56.16GNUtooouch....
13:56.26GNUtoosorry for bothering then
13:56.29cr2_NetRipper: irq should not be short
13:56.56cr2_NetRipper: the irq numbers are more or less hardcoded
13:57.13cr2_#defines
13:57.26NetRipperwell the only way for a -1 to be 65535 is when it's a unsigned short
13:57.28NetRipperright?
13:57.28cr2_offset by MSM, and SIRC maybe
13:57.37cr2_NetRipper: yes
13:57.40CaptnoordNetRipper: yup
13:57.50Captnoord0xFFFF == -1
13:57.59NetRipperhence my blabbering about conversion problems
13:58.01Captnoord0xFFFE == -2
13:58.10NetRipperi may be off ofcourse but still
13:58.13*** join/#htc-linux ulaas (~ulaas@88.253.0.156)
13:58.17NetRipperworth looking into i suppose
13:58.44cr2_NetRipper: i'm a bit worried about the smsm irq area
13:58.54cr2_if we have it at all
13:59.14cr2_i have added the Leo_SMEM page, we need to put a table there
13:59.25Markinuscr2_: ok, I needed a bit time but now I have the same values too :P
13:59.26cr2_with names from smd_private.h
13:59.36cr2_and actual smem data dump
13:59.47Markinuscr2_: du we have MDP22 oder 32?
13:59.49Markinusdo
13:59.58Markinussry, 31
14:00.07cr2_Markinus: ok, so we know how to fix mdp_lcdc now, and also the uboot lcd setup
14:00.12cr2_Markinus: 31
14:00.14NetRipperanyway i gtg, talk to you guys later ;)
14:00.23cr2_NetRipper: ok
14:01.14cr2_Markinus: the biggest problem with uboot is that we can't put it at the same area in SMI as the current code
14:01.44cr2_or maybe we can ??
14:02.36Markinuscr2_: I I see this right, then we cave to flash uboot to the device right?
14:02.41MarkinusIf
14:02.54*** join/#htc-linux M1DLGpc (~M1DLGpc@bb-87-81-252-83.ukonline.co.uk)
14:03.02*** join/#htc-linux Neo31 (~Neo31@unaffiliated/neo31)
14:03.11*** part/#htc-linux ulaas (~ulaas@88.253.0.156)
14:03.54cr2_Markinus: if it will work properly, you can use it instead of spl
14:04.27cr2_Markinus: but then you will loose wince compatibility
14:04.33Markinusok, the we can boot wince too, right?
14:04.37Markinusahh ok
14:04.41Markinushmm
14:04.43Markinusnot good :)
14:05.15cr2_it's better to boot uboot from haret, or as a biag.nbh
14:05.18Markinuscr2_: I thought we can boot wince too . . .hmmm
14:05.22cr2_s/biag/diag/
14:05.30Markinusif this is working, yes :)
14:06.38cr2_where is uboot loaded in smi ?
14:06.40Markinuscr2: so we have to modify the QSD8_x50_surf.h . . .
14:07.12cr2_Markinus: the ebi banks probably too
14:08.29MarkinusI changed the PHYS_SDRAM_1 to the 0x11800000 from Leo . .
14:09.27*** join/#htc-linux [1]Captnoord (~Captnoord@dc5147a47b.adsl.wanadoo.nl)
14:10.57*** join/#htc-linux k_linux (~klinux@41.248.145.180)
14:11.11cr2_Markinus: yes, you can still use 2 banks
14:11.32cr2_Markinus: check where is the SD card, if it is on the right controller
14:11.46cr2_and if the gpio alt settings are the same
14:11.59k_linuxhi guys
14:13.20k_linuxcan someone help me ? I'm working on esspresso 2.1 android until now I have sound data gsm wifi working
14:13.58k_linuxI need help for BT and Gsensor
14:14.30cr2_Markinus: +#define UBOOT_SDRAM_BASE         0x00000000      /* SMI */
14:14.43cr2_Markinus: we may move it to a known working area
14:15.00cr2_Markinus: the nand values should be checked too
14:15.03cr2_+// NAND device specific register values for the NAND controller
14:15.04cr2_+// These values are for the Samsung MFG=0xEC DEV=0xAA device (x8, 1.65~1.95V, 2K page)
14:15.51cr2_+#define SDC_INSTANCE  1
14:16.07cr2_should be 2 ? 1 is wifi afair
14:17.02Markinuscr2_: where is the second bank on leo . .I can'T see it in the header file . .
14:17.20cr2_Markinus: it's contiguous
14:17.30cr2_Markinus: but you can set it to 0x20000000
14:17.46cr2_and simply split the ram into 2 pars
14:18.29Markinusahh, yes :)
14:18.46k_linuxhowto dump log which is on the first boot? logcat begins after android boots
14:19.16cr2_+#define SMEM_START  0x00100000
14:19.19cr2_it's ok
14:20.01leviathanGNUtoo: sry, I was eating something
14:20.07GNUtoonp
14:20.22GNUtooI started importing the new driver btw
14:20.32GNUtooI won't commit it now tough
14:20.34Markinus#define CONFIG_NR_DRAM_BANKS    2   /* There is 1 bank of SDRAM */
14:20.34Markinus#define PHYS_SDRAM_1            0x11800000     /* EBI1 */
14:20.34Markinus#define PHYS_SDRAM_1_SIZE       0x7400000
14:20.34Markinus#define PHYS_SDRAM_2            0x20000000     /* EBI2 */
14:20.34Markinus#define PHYS_SDRAM_2_SIZE       0x11800000
14:21.56cr2_+#define SMEM_START  0x00100000ok
14:22.21cr2_Markinus: +#define LINUX_MACH_TYPEMACH_QSD8X50_SURF
14:22.23Markinuscr2_:  SMEM_START or  UBOOT_SDRAM_BASE ?? Secod one or?
14:22.35cr2_Markinus: smem is ok.
14:22.49cr2_Markinus: need to think about smi
14:23.28leviathanGNUtoo: does it work then? I mean your "new" driver
14:23.53GNUtoonot yet I'm working on it
14:24.16GNUtooIt doesn't compile for now
14:24.51Markinuscr2_: the mtype is already set correctly, SMEM in nand_qc.c too
14:27.43*** join/#htc-linux Vito89 (~quassel@gw.loccal.net)
14:27.58*** join/#htc-linux wout (~5070d0f6@gateway/web/freenode/x-ngkoberstfwmmbwe)
14:28.06wouthey guys
14:28.20woutis there a way to output a list of all 2g/3g switch events from the terminal?
14:28.30Markinuscr2_: and you meant something about to correct the 24Bit Color patch
14:28.52cr2_Markinus: we have 16bit lcd
14:29.07phhwout: maybe with logcat -b radio
14:29.16cr2_Markinus: and the MDP/LCDC registers need to be adjusted for our dumped values
14:29.22Markinusahh, yes . .wince cannot more . .
14:29.41phhMarkinus: really ? wow
14:29.50Markinusyeah.....
14:29.59woutdoes that work in a similar way to dmesg?
14:30.08phhwout: what do you mean ?
14:30.33phhI love windows desktop.
14:30.36cr2_Markinus: we need to dump the nand config registers too
14:30.41woutwell, I used dmesg to get a list of enter/exit sleep activity, would like something similar for 2g/3g
14:30.47phhI can't even install HTC Sync without a bug ...
14:30.49woutto see how it relates to my battery uptime
14:30.58phhwout: hum no timing in it I think :/
14:31.53phhah and maybe it doesn't give the changes when in sleep
14:32.06Markinuscr2_: this one: ?  0xa0a00000   2   NAND+MPU
14:32.11woutah, ok
14:32.23woutback to work, later
14:32.33cr2_Markinus: yes, the CFG regs
14:33.59k_linuxhi phh
14:34.20phhhi k_linux
14:34.56k_linuxdid u see my question about 2.1 espresso android ??
14:35.21k_linuxin fact I have flash player 10.1 working !!
14:35.33phhno
14:35.52phhand espresso ?
14:35.55k_linux(14:13:18) k_linux: can someone help me ? I'm working on esspresso 2.1 android until now I have sound data gsm wifi working
14:35.55k_linux(14:13:55) k_linux: I need help for BT and Gsensor
14:36.22k_linux2.1 android not eclair
14:36.24phhand the gui works ?!?
14:36.36k_linuxyes but undersized
14:36.38phh.
14:36.40phhok.
14:37.04k_linuxI ripped all htc's widgets and it's working like a charm !
14:37.16phhbut you won't be able to do anything with a undersized screen ...
14:37.29phhand you won't be able to "extend" it because of lame 3D hardware
14:37.34k_linuxno without rosie everything is ok
14:37.44Markinuscr2_: in wich header file are the  CFG regs defines? .. . .searching . .
14:37.48phhk_linux: even the widgets ?
14:38.12k_linuxyes they are working
14:38.25phhweird
14:38.33k_linuxdo u want to try it ?? :)
14:38.35phhany, for gsensors, just put the system/lib/hw/sensors.default.so
14:38.48k_linuxok will try it now
14:38.54phhand for BT, edit the init.rc, see last stinebd's commit
14:38.58phhand yes I want to try
14:39.14k_linuxok I'll share it somewhere
14:39.14phhbut right now i'm on windows trying to make htc sync working. but it won't even install. I love windows.
14:40.00k_linux"ah bon depuis qd ?? :)"
14:40.15phhwhat ?
14:40.45k_linuxsorry I know that all your machines are under linux
14:40.48cr2_Markinus: i'll be back in an hour
14:40.55Markinuscr2_: ok
14:40.59k_linuxwhen u change it ??
14:41.50Captnoordrelax
14:41.52Captnoordhave patience
14:47.09k_linuxphh: I forgot something 2.1 is not as flawless on blac as on diam u know the reason !
14:50.28phhk_linux: ???
14:50.37phhI'd expect it's the exact opposite -_-'
14:51.26Captnoordhmm so blac just doesn't have enough memory for gpu?
14:51.52k_linuxI'll explain, 2.1 need a lot of memory, so it works on blac as eclair on diam !
14:52.03Captnoordlol
14:52.04phhCaptnoord: depends on what you do with it
14:52.05Captnoordthat I know
14:52.11Captnoordhmmm...
14:52.13phhk_linux: diam has only 196MB ....
14:52.26phh188 ?
14:52.29phhwell 128+64.
14:52.43Captnoordso one reason why blac doesn't work smooth is because of the memory layout
14:52.56k_linuxphh: yes but what is used is 128 vs 188 on blac
14:53.07GNUtoostrange strange...
14:53.09GNUtooarch/arm/mm/init.c:353: error: expected identifier or '(' before 'unsigned'
14:53.12phhk_linux: ?!?
14:53.19phhk_linux: you're using a damn old kernel then
14:53.22GNUtooint pfn_valid( unsigned long pfn ){
14:53.26phhuh ?
14:53.28phhno sorry.
14:53.30phhi misread
14:53.36k_linuxphh: ok
14:53.36phhk_linux: well it uses all memory available
14:53.38phhnothing new here
14:55.50*** join/#htc-linux babijoee (~babijoee4@d110-33-179-143.sun801.vic.optusnet.com.au)
14:56.18*** join/#htc-linux balans (~Gebruiker@82-170-215-5.ip.telfort.nl)
14:57.00phhk_linux: maybe hw3d is disabled on your blac but not on your diam ?
14:57.06phhbut that sounds really strange
14:58.59k_linuxphh: Gsensor is OK !! thx
14:59.15GNUtoommm maybe it's already defined in mmzone.h
14:59.25phhk_linux: oh I'm just thinking
14:59.27*** join/#htc-linux Zoolooc (~fredsibar@p54957F9F.dip.t-dialin.net)
14:59.32k_linuxit's enabled
15:00.17phhk_linux: where do you get this rom from ?
15:00.18phhdirectly from an android device ?
15:00.18phhor from a vogue tree  ?
15:00.18k_linuxphh: kaiser's forum
15:00.18phhhum
15:00.18phhoh well it's still worth a try
15:00.30phhfor bluetooth there might be some things interesting
15:00.38phhis there a btilp in system/bin?
15:01.18k_linuxphh: no
15:01.23Captnoordhmm the wince batt driver is working in Coulomb I think
15:02.14phhCaptnoord: but they don't have precise current measurement ?!?
15:02.36phhk_linux: can you pastebin the list of the apps in system/bin ?
15:02.39Captnoordbattery.charge_full_time_diff
15:02.48Captnoordhas a if max check of 4000
15:02.53Captnoordand if its higher than 4000
15:02.56Captnoordits set to 3600
15:02.59k_linuxok
15:03.03Captnoordwhich is the exact load of 1A/h
15:03.13Captnoord1000 mAh
15:03.17phhCaptnoord: 1Ah :p
15:03.21phh.
15:04.45k_linuxphh: http://pastebin.com/a5zQZbD6
15:05.06phhk_linux: bluetooth works on kaiser ?
15:05.34k_linuxphh: don't know
15:06.23k_linuxphh: the link of original rom http://forum.xda-developers.com/showthread.php?t=613826
15:06.31GNUtoothere is no pfn_valid in codeaurora kernel
15:06.57GNUtooin init.c in arch/arm/mm
15:07.12Captnoordalright
15:07.19babijoeecr2: ping
15:07.25Captnoordthey use a 0.1 ohm current measure resistor
15:07.36Captnoordthats pretty normal
15:07.53phhk_linux: you'd better start from a hero's system ....
15:08.22phhwell, the original espresso rom.
15:08.47k_linuxphh: ok
15:08.58phhunless you're really lazy.
15:09.13phhconcerning bluetooth, btilpd should work
15:09.20phhand that would make a lot of people really happy :p
15:12.22phh(maybe not replicant people though.)
15:13.40k_linuxphh: btilpd ? I don't have it neither on eclair nor on sense
15:13.45k_linuxcan u share
15:13.48k_linuxit ?
15:14.41phhk_linux: read better ...
15:14.46phhit's in the pastebin you gave me ..
15:16.35k_linuxphh: no it's not the case what I have is btipsd 
15:17.24phhoh yeah right.
15:17.29phhtypo.
15:18.25phhor there are two bt ti stacks ?
15:18.27phhdon't remember -_-'
15:19.30k_linuxphh: np
15:19.44k_linuxphh: even eclair don't have it
15:19.56k_linuxphh: BT works on it
15:20.19phhk_linux: call on BT works maybe ?
15:21.20k_linuxphh: of course no, I mean it can be enabled & pairs with other devices
15:22.04phhthere is no btipsd reference in the init.rc ?
15:23.29babijoeephh: this isnt really important but whats the stock battery capacity shipped with your diamond?
15:23.40phh900mA
15:23.41k_linuxphh: no
15:23.50babijoeeooo
15:24.04k_linuxphh: should I add it somewhere ??
15:24.09babijoeenow u have 1800 right?
15:24.14phhbabijoee: 1500
15:24.16GNUtoobtw leviathan :
15:24.20k_linuxbabijoee: hi
15:24.20GNUtoodrivers/mmc/host/msm_sdcc.c: In function 'msmsdcc_probe':
15:24.20GNUtoodrivers/mmc/host/msm_sdcc.c:1164: warning: passing argument 2 of 'request_irq' from incompatible pointer type
15:24.20GNUtooinclude/linux/interrupt.h:117: note: expected 'irq_handler_t' but argument is of type 'int (*)(int,  void *)'
15:24.23leviathanyes?
15:24.25phhbabijoee: hum well, on rhod or diam ?
15:24.36phhk_linux: give me your current build, I'll see
15:24.44phhit depends on how much kaiser guys broke it
15:24.45leviathanhmmm
15:24.57k_linuxphh: will upload it now
15:25.12leviathanGNUtoo: possible that the error we're encountering with wifi comes from there
15:25.13leviathanyes
15:25.20GNUtoook
15:25.34babijoeehi k_linux
15:25.49babijoeediam
15:25.57babijoeejust checking
15:25.57*** join/#htc-linux Zoolooc_ (~fredsibar@p54956E4D.dip.t-dialin.net)
15:26.16babijoeeit sounded weird when i saw someone with extended battery of 1100mah on diamond
15:26.28babijoeei though normal standard was 1400mah
15:26.31k_linuxbabijoee: u know I have 2.1 working on blac with 3G gsm Gsensor sound wifi !!
15:27.02babijoeeno i didnt
15:27.31babijoeeand?
15:28.22babijoeephh: have you tried new build?
15:28.29phhbabijoee: no
15:28.39babijoeepeople on forums keep complaining about downloading from market problems
15:28.41babijoee=/
15:28.44phhmy extented diam battery is 2Ah
15:28.54babijoeeah k
15:29.04phhbabijoee: don't know your changes
15:29.11babijoeeit must be as fat as raph
15:29.32phhas fat as rhodium
15:29.43babijoeetopaz sensor and some apps irrc
15:29.46babijoeelol
15:30.37babijoeeyeah according to stinebd launcher2 requires some work with gl
15:31.02babijoeebut i've added home++ which is pretty good to me
15:31.04phhdo we have the sources of launcher2 ?
15:31.11babijoeehmmm
15:31.16babijoeei haven't looked
15:31.21babijoeebut dont think so
15:31.29phhok
15:31.41babijoeehome++ is much better imo
15:31.51phhah. ok
15:31.55phhno need to work on that then :p
15:32.14makkonenI don't like home++ not having the notifications bar displayed. otherwise, it's a big improvement.
15:32.26babijoeemakkonen: me too
15:32.29phhmakkonen: I think it's a setting, no ?
15:32.30Markinusarghhh .. I need more RAM . . 4 GB are to less .  .
15:32.30*** join/#htc-linux miknix (~miknix@gentoo/developer/miknix)
15:32.36babijoeetook me awhile to figure that one
15:32.45makkonenphh: it might be. I didn't look closely enough to find it, though.
15:32.55phhbabijoee: just read the first start page ! :p
15:33.02phhmakkonen: ok
15:33.04babijoeenah
15:33.13babijoeei never read those blurbs
15:33.14babijoee-.-
15:33.21phhbabijoee: me neither.
15:33.28phhbut when I've seen the homepage
15:33.35phhI decided I'll change my mind once.
15:33.39babijoeelol
15:33.53babijoeeits pretty straight forward if you've used android for ahiwle
15:33.57babijoeewhile*
15:35.02makkonenI was using google reader in the android browser this morning, catching up on blog posts... it works SO WELL. I think it was the first time I actually just used my phone in android.
15:35.30babijoeeanways i'll release an updated build tomorrow, i deleted gtalk service and removed htc_ime from apps in /system/apps
15:35.55phhmakkonen: I hate google reader. No push-rss.
15:36.37makkonenah, that's a shame. I was just reveling at the ajax-y expand and collapse of articles.
15:36.54phhinstead I use rss2email :/
15:37.04phhit's really not as great as a real rss reader
15:37.24makkonenI quite liked rsshub in winmo.
15:41.27phhchamonix: that makes me think. if you want to make a really great app. just publish your rss2email + tagging stuff to the world.
15:44.54*** join/#htc-linux dcordes (~dcordes@unaffiliated/dcordes)
15:45.01dcordeshi
15:45.06phhok. i definitely hate windows.
15:45.28dcordesleviathan are you still here?
15:46.13Markinusphh: here is a libaudiofinger cpompiled with A2DP for testing: http://www.megaupload.com/?d=M288IG8W
15:46.34phhstill no A2DP headset uh ?
15:46.49MarkinusNo :(
15:48.23phhk_linux: still not ?
15:48.32*** join/#htc-linux [1]Captnoord (~Captnoord@dc5147a47b.adsl.wanadoo.nl)
15:49.15k_linuxphh: I have a pb with firefox I should restart on windows to upload the files
15:49.25phhtss ...
15:50.22Markinusphh: do you have one for a test?
15:50.33phhMarkinus: yes, but the android on my diam is quite old
15:50.35phhI have to update it first
15:51.04Markinusphh: ok, would be great to know . ..
15:51.46*** join/#htc-linux bzo (~chatzilla@adsl-75-35-5-163.dsl.pltn13.sbcglobal.net)
15:51.48Markinusvbox is eating RAM . . arghhhh. . . after 3 days 2 GB on RAM . .very hungry. .
15:51.59phhreboot :p
15:52.04Markinusyes . . indeed.  .
15:52.32makkonen~usb_function_switch
15:52.33aptrumour has it, usb_function_switch is /sys/devices/platform/msm_hsusb/usb_function_switch
15:52.47phhmakkonen: do you know if htc sync works with setting it to 2?
15:53.28makkonenI do not know. never tried htc sync
15:54.41phhok
15:56.21bzophh: emailed you the acpuclock merge patch
15:56.30phhbzo: ah cool thanks
15:56.53bzoso actually I was wrong about codeaurora version using freq tables
15:57.07bzoit does use them, but it dynammically detects the pll freqs and chooses the right one
15:57.32bzoand it does create the cpufreq table from scratch
15:58.10phhok
15:58.45bzoyou may have to tweak the freq tables a little for the turbo mode, I don't have one to test
15:59.09bzoI had to make a few changes to the regular one for the cpu freq changing to work well with our config
15:59.30bzomainly, our config specifies a min freq of 128000, while the tables assume you can go lower than that
16:00.12phhthe problem is that we don't have perflock
16:00.16phhso we can't set min freq to 0
16:00.39bzook
16:00.56bzophh: do you know why the pll on/off control was never implemented?
16:01.14phhbzo: it's not really useful
16:01.16k_linuxphh: uploading
16:01.25phhk_linux: waiting.
16:01.33bzoit will not save power in suspend?
16:01.40phhnot much
16:01.52bzook, I won't worry about it then
16:02.06ToAsTcfhbzo : what u found for turbo mode?
16:02.16phhwhenever we get <10mA total power consumption, we might start worring about it.
16:02.47ToAsTcfhi have turbo mode :)
16:02.50bzoToAsTcfh: nothing really. We are going to a newer version of acpuclock that supports lots of different configs
16:02.59ToAsTcfhoh
16:03.09Captnoordokey lets dump some battery data
16:03.18Captnoorddamn... I can't believe our driver has so much crap in it
16:03.22Captnoordbleh
16:03.24phhhaha
16:03.35bzoToAsTcfh: probably not worth merging into your kernel if you only support one device
16:03.41CaptnoordI mean... I can understand some filtering or avaraging..... but this
16:03.42Captnoordbleh
16:03.52*** join/#htc-linux the_sys0p (~the_sys0p@cpe-67-49-210-192.bak.res.rr.com)
16:04.07phhbzo: one day we will binary support all msm7*a devices ! :p
16:04.45ToAsTcfhwell i still get kernel panics when i set the config max equal or higher then its set in acpuclock. but if i set them to 528 thereis no issue
16:04.46bzophh: not all msm* devices? hehe :P
16:04.54phhbzo: that will be tougher to do :p
16:05.36bzoToAsTcfh: in your dmesg near the beginning, it shows the PLL freqs. Do they actually change when you oc?
16:06.12*** join/#htc-linux miknix (~miknix@gentoo/developer/miknix)
16:06.27ToAsTcfhwell i cant over clock. it changes the max in cpu0 but i casnt jump up to pll2
16:06.31Captnoordbump batt info every second
16:07.31ToAsTcfhin other words if i can get it to boot itll set the max but i can never ramp up to it. im stuck on pll1 480
16:08.12bzoToAsTcfh: right, but it is a different thing to check if the PLL is actually getting configured
16:08.55ToAsTcfhwhat u mean? it is actually setting the min and max on the device
16:09.13ToAsTcfh[    5.422698] msm_cpufreq_target 595200 r 1 (245760-595200) selected 595200
16:10.14ToAsTcfhbut idk if the pll is being configured for that or if its just a max being set
16:11.04ToAsTcfhsuggestions for checking?
16:12.05bzoToAsTcfh: I mean near the beginning of the dmesg, if prints all the base PLL freqs
16:12.57ToAsTcfh[    2.661865] OC: ADJUSTING FREQ TABLE freq=198400 div=2 ahbclk=99200 ahbdiv=1
16:12.57ToAsTcfh[    2.661895] OC: ADJUSTING FREQ TABLE freq=297600 div=1 ahbclk=74400 ahbdiv=3
16:12.57ToAsTcfh[    2.661926] OC: ADJUST freq_tbl_entry: 528000 to 595200
16:12.57ToAsTcfh[    2.661956] OC: ADJUSTING FREQ TABLE freq=595200 div=0 ahbclk=148800 ahbdiv=3
16:12.57ToAsTcfh[    2.662628] ACPU running at 480000 KHz
16:12.58ToAsTcfh[    2.673492] msm_i2c_probe
16:13.00ToAsTcfh[    2.673736] msm_i2c_probe: clk_ctl 35d, 100000 Hz
16:13.27*** join/#htc-linux Unholy (~IceChat7@24.227.27.77)
16:13.42The-CompilerToAsTcfh: http://en.wikipedia.org/wiki/Pastebin
16:15.14ToAsTcfhhttp://en.pastebin.ca/1813615
16:15.34ToAsTcfh<The-Compiler> ToAsTcfh: http://en.wikipedia.org/wiki/Pastebin <<~~~ sorry
16:16.38Captnoordlol
16:16.39ToAsTcfhbzo that dmesg is from one that wont boot
16:16.53The-CompilerI don't really care, but others might do ;)
16:17.09ToAsTcfhok
16:17.18ToAsTcfhsorry everyone :P
16:18.23bzoToAsTcfh: your kernel doesn't print out the PLLs settings like ours, but the problem is not obvious to me
16:18.42ToAsTcfhthatsux
16:19.20ToAsTcfhyeah its strange because when it does boot the dmesg looks the same
16:19.30Captnoordgoes aaaaahhhh crap... didn't enable ram consoel
16:19.32Captnoordconsole
16:22.39phhCaptnoord: haha
16:25.13Captnoordlol
16:25.23Captnoordnow every second the batt level falls
16:25.31Captnoordfluxuate majorly
16:25.36Captnoordbetween 36 and 45
16:25.37Captnoordlolz
16:28.24leviathandcordes: I made a cofee
16:28.30leviathansmoked some cigarettes
16:28.39leviathanand tried to get away my hadace
16:28.58leviathanheadache
16:29.24leviathanhere we have high pressured air, perhaps rain will follow soon
16:30.02phhk_linux: ?
16:30.21leviathandcordes: after I didnt manage to get the build errors away I decided to rebuild everything from scratch in bb
16:30.37leviathanuntil its finished I guess I will solve my working sheets for study
16:30.39leviathanfor next week
16:30.43k_linuxphh: yes still uploading my connection is too slow !
16:31.00k_linuxphh: still 18 min
16:31.10phhok
16:31.38k_linuxphh: I'll try a new system.sqsh from hero 2.1 now
16:33.52phhwhy ?
16:33.59phhah because I told you to.
16:34.06k_linuxphh: :)
16:38.35k_linuxphh: I have pb with latest kernel in blac it's about galloc and out of memory is it hw3d ??
16:39.05phhblac hasn't enough memory to handle splashscreen through 3D
16:39.59k_linuxphh: ok so if I remove splashscreen it will work ?
16:40.06phhI think so
16:40.08Captnoordphh maybe we can make it use less memory by forcing gl es with diff config
16:40.30phhCaptnoord: I want to force it in lower resolution,  but no clue how :
16:40.31phh(
16:40.45Captnoordnah you don't want the lower resolution
16:40.55Captnoordhow much mem gl needs
16:40.55phhuh ?
16:41.02phhhow do you want to do that ?
16:41.09Captnoordis related to how mutch deph buffer memory is allocated
16:41.17Captnoordif you allocate 16 bits or 32 bits memory for that
16:41.20Captnoordyour gonna have problem
16:41.21Captnoords
16:41.21*** join/#htc-linux mcdull (~blackston@061093132071.ctinets.com)
16:41.28phhah.
16:41.28Captnoordespecialy when you don't have anough
16:41.28*** join/#htc-linux Markinus (~Miranda@gtng-4db048e3.pool.mediaWays.net)
16:41.29Captnoordso
16:41.34phhbut we have no control on it
16:41.38Captnoordnot?
16:41.45phhwell, for bootsplash maybe, we have the source
16:41.48phhbut not for most app
16:42.05phhunless I missed something
16:42.06Captnoordhmmm also a problem can be is how the color map of the device is set
16:42.23Captnoorddoes it use 32 bits
16:42.24Captnoordor
16:42.27Captnoord24 bits color map
16:42.40Captnoordbut yea
16:42.50Captnoordother apps might cause problems
16:42.51Captnoordhmmmm
16:43.10phhthe only solution I can see is scalnig
16:43.34Captnoordsmem is 2 bytes alligned
16:43.43Captnoordv = (values_32[2] * 9 / 7) + (values_32[4] / 7) - (values_32[3] / 28);
16:43.50Captnoordthat is completely useless
16:43.54Captnoordand the only reason it works
16:44.04Captnoordis because the value's 4 is empty
16:44.08Captnoordbleh
16:44.17phh:D
16:44.39Captnoordnow changed my dump stuff to 16 bits
16:44.47makkonenwell, it works because I tested out numbers until it worked. :-p the discharge value is just empty, though. I don't know what the story is on that.
16:45.04phhi'm going to kill k_linux's connection.
16:45.06Captnoordthe discharge value isn't empty
16:45.07Captnoord:P
16:45.19phhmakkonen: it's empty maybe when charging ? :p
16:45.20*** join/#htc-linux tehtrk (~tehtrk@rrcs-24-173-220-30.sw.biz.rr.com)
16:45.21Captnoord[  188.565704] batt_id: 221184007
16:45.22Captnoord[  188.565765] batt_vol: 1510264
16:45.22Captnoord[  188.565795] batt_vol_last: 1
16:45.22Captnoord[  188.565826] batt_temp: 0
16:45.22Captnoord[  188.565856] batt_current: 0
16:45.22Captnoord[  188.565887] batt_current_last: 0
16:45.22Captnoord[  188.565917] batt_discharge_current: 4
16:45.27makkonenok, then... it isn't values_32[4] then
16:45.29Captnoordthats with fucked up allignment
16:45.32Captnoordnah
16:45.36Captnoordits value_16
16:45.38Captnoordnot 32
16:45.44Captnoordon my raph it is....
16:46.00Captnoord221184007 = D2F0007
16:46.07makkonenwell, the battery values are pretty different on diam500/raph800 and the rest, from what I've seen.
16:46.08Captnoordbatt_id = 0xD2F
16:46.13Captnoordbatt_vol == 7
16:46.38Captnoordbatt_vol = 170B78
16:46.48Captnoordvol_last is 0x170
16:46.54Captnoordbatt temp is 0xb78
16:47.00Captnoordnah
16:47.04Captnoordfuck the hard shit
16:47.09Captnoordas I changed it anyway
16:47.11makkonenso has it just been done wrong the whole time?
16:47.19Captnoordyup
16:47.26makkonenfunny
16:47.30Captnoordor
16:47.36Captnoordat least on my raph
16:47.55Captnoordbut not only that
16:48.06Captnoordits full of crap
16:48.07CaptnoordI mean
16:48.08makkonendoesn't yours use the values_16 section, not _32, anyway?
16:48.15makkonenoh, it's terrible. across the board.
16:48.17Captnoordatm now it uses 16
16:48.29phhCaptnoord: you pasted _32
16:48.30Captnoordgonna dump ram console
16:48.33Captnoordto see
16:48.34Captnoordyup
16:48.37CaptnoordIdid
16:48.50CaptnoordI need to dump the 16
16:49.05phhwait, you have _32 AND _16 in your smem ?
16:49.16Captnoordwait
16:49.24Captnoordi'll explain when i've read my ram console
16:49.28phhok
16:50.10Captnoord[  106.589691] batt_id: 7
16:50.10Captnoord[  106.589691] batt_vol: 3426
16:50.11Captnoord[  106.589721] batt_vol_last: 2933
16:50.11Captnoord[  106.589721] batt_temp: 23
16:50.11Captnoord[  106.589721] batt_current: 1
16:50.11Captnoord[  106.589721] batt_current_last: 0
16:50.11Captnoord[  106.589752] batt_discharge_current: 0
16:50.12Captnoord[  106.589752] VREF_2: 0
16:50.12Captnoord[  106.589752] VREF: 0
16:50.13Captnoord[  106.589752] ADC4096_VREF: 0
16:50.13Captnoordthats better
16:50.22phhmakkonen: just thinking, have you tried comparing hero's smem stuff to yours ?
16:50.25Captnoord23 degrease is kinda correct
16:50.26phhmakkonen: it might be quite the same thing
16:50.41Captnoordi'm using hero's dump stuff atm
16:50.50phhCaptnoord: oO
16:50.55phhbut hero has 32bit integers
16:50.56phhwe have 16
16:50.59phh(according to doc.)
16:51.01Captnoordyup
16:51.11Captnoordbut I can change the var types of the struct
16:51.14Captnoord:P
16:51.24phhyes but if the size is wrong, it's likely that everything is wrong
16:51.27Captnoord#define ubatt u16
16:51.27Captnoord#define sbatt s16
16:51.27Captnoordstruct htc_batt_info_full {
16:51.27Captnoordvolatile ubatt batt_id;
16:51.27Captnoordvolatile ubatt batt_vol;
16:51.27Captnoordvolatile ubatt batt_vol_last;
16:51.28Captnoordvolatile ubatt batt_temp;
16:51.28Captnoordvolatile ubatt batt_current;
16:51.29Captnoordvolatile ubatt batt_current_last;
16:51.30*** join/#htc-linux miknix (~miknix@gentoo/developer/miknix)
16:51.37Captnoordthat is what I had
16:51.42Captnoordthe wrong stuff
16:51.46babijoeegnite guys
16:51.48Captnoordnow i've changed it to u16
16:51.50Captnoordand its correct
16:52.05Captnoordbattery is low...
16:52.18Captnoordbatt_vol: 3426... its about correct
16:52.19phhhave you tried the space area cdma uses ?
16:52.24phhI wonder if it could be right.
16:52.32Captnoordnope
16:52.43phhCaptnoord: there is a not really constant +0.3V delta
16:52.54phhmaybe linked with vref and vref/2
16:53.19Captnoord[  106.589813] pd_temp: 2850
16:53.33Captnoordvref isn't used where I dumped the value's
16:53.39CaptnoordI wonder if thats related to the dex call
16:54.01phhbtw there are two battery dex calls
16:54.25Captnoord139 == get battery id
16:55.01Captnoordand 138 for get data
16:57.51Captnoordlol
16:58.37*** join/#htc-linux BabelO (~fcr@unaffiliated/babelo)
16:58.48Captnoordstrange
16:58.52Captnoordwhile .smem_field_size = 4, is set
16:58.54Captnoordfor raphael
16:59.25phh?!?
16:59.27Markinusargh.... a whole cup coffee in my keyboard . .I love this . .
16:59.43phhMarkinus: my phone won't peer with my headset ...
16:59.53phhdon't know what's the problem :/
17:00.05phhbut people reported they could peer with their headset even without the lib
17:00.09phhso there is something wrong on my side
17:00.24Captnoordhmmmm
17:00.27Captnoordwierd stuff
17:00.32Captnoordi've been reading the wrong shit
17:00.34Captnoordall along
17:00.36Captnoordnice
17:00.55phhCaptnoord: no way you have field_size=4
17:01.02adamwmorning
17:01.06Captnoordyea
17:01.09Captnoordstrange shit
17:01.10Captnoord:S
17:01.10phhadamw: afternoon.
17:01.19CaptnoordI had to change to u16 otherwise I would read crap
17:01.22Captnoordhmmm
17:01.22Captnoordwait
17:01.28Captnoordmaybe what i read
17:01.28ToAsTcfhbah
17:01.29Markinusphh: hmm, ok. Thx for testing. .  .  maybe I do domething worng too. . I have only 2.1 repo . .maybe it'S a problem with 2.01. ..
17:01.30Captnoordis wince shit
17:02.01phhMarkinus: na it's not related to the libaudioflinger.so you gave me
17:02.55Captnoordi'm having a hard time understanding why the batt driver kinda worked
17:02.56Captnoord:S
17:03.07Captnoordthe way I see it
17:03.12Captnoordit shouldn't even work
17:03.18makkonenphh: I didn't realize hero had any battery smem stuff worth looking at. I'll check it out.
17:03.42Captnoordi've changed the default battery offset to the correct one
17:04.04phhmakkonen: and the size matches
17:04.11makkonencool
17:04.26phhmakkonen: iirc it has circa 40 entries, and there is place for 48 :p
17:05.25phhmakkonen: you have the sources of it or you want me to paste it ?
17:06.54makkonenis it in the heroc kernel source?
17:07.01phhin htc_battery.c
17:07.10makkonenok, that's where I'm looking now.
17:08.23phhdump int @ 0x01f00000+0xfc140+19*4
17:08.26phhit should be a percentage
17:08.34phhand 18.
17:09.02*** join/#htc-linux Wout (~wout85@s55927d21.adsl.wanadoo.nl)
17:09.37Woutphh, was in an area with less 3g switching today, 54% left after 6 hours, so that extrapolates to 12hrs total
17:09.38Woutnot bad
17:09.52Woutmostly standby though
17:09.57phhit means you was still switching ?
17:10.04phhwhere do you live to switch ? -_-'
17:10.15phhthe only place where it's not constant is my school
17:10.28Woutwork in a store, and the reception in the stock room is pretty bad
17:10.38Woutbut was in the store itself for most of the time today
17:11.03*** join/#htc-linux balsat (~balsat@87.72.13.34)
17:11.30Captnoordokey
17:11.32Woutthats why I asked about the logging thing, was interested to see how much it switched between 2g/3g today
17:11.35Captnoordlets make things a bit worse
17:11.38Captnoordwince batt driver
17:11.45Captnoordreads its stuff after the dex call from
17:11.45Captnoord0xAA0FC110
17:12.03phh(NB: x200-x140 != x60.)
17:12.14Captnoordhehe
17:12.17phhCaptnoord: that's where we read it
17:12.21Captnoordreally?
17:12.25phh-_-'
17:12.30Captnoord.smem_offset = 0xfc140,
17:12.31phhhttp://htc-linux.org/wiki/index.php?title=RaphaelMemoryMap
17:12.35CaptnoordI know
17:12.43Captnoordsmem_offset of the raphael
17:12.51Captnoordso it should be
17:12.53phhno.
17:12.56Captnoord0xFC110
17:13.00phhhtcraphael_device_specific_fixes
17:13.02phhread that.
17:13.06phhthis function i mean
17:13.39Captnoordmsm_htc_hw_pdata.battery_smem_offset = 0xfc110;
17:13.39Captnoordmsm_htc_hw_pdata.battery_smem_field_size = 2;
17:13.41Captnoordalright
17:13.50Captnoordpoints at size = 2
17:14.08Captnoordwhile we are reading 4
17:14.18makkonen0xfc140, size = 4 is the raph800/diam500 location
17:14.55Captnoordmakkonen: i didn't know about the fix part
17:15.09makkonenwell, just looked over the heroc htc_battery.c file, and remembered that I don't really know C.
17:15.48makkonenit references RPC calls for set_delta and set_full_level, which seems interesting.
17:16.02phhmakkonen: we don't have those RPC
17:16.04phhI don't think you have either
17:16.20phhmakkonen: see /dev/oncpc if you have a 301xxxx thing
17:21.43makkonenso how does it get done in wince?
17:22.08Captnoordsmem
17:22.34phh+dex call
17:22.39Captnoordyup
17:23.17CaptnoordI think ive had some stuff around
17:24.45bzoCaptnoord: did you ever find anything useful disassembling the battery dll?
17:25.44Captnoordyea but as i'm not really good at it
17:25.52CaptnoordI didn't really had time to use it
17:25.58Captnoordbut now its a good time to try
17:27.27CaptnoordI think that get_batt_id does more than its name suggests
17:28.23bzohaha, well for one it seems to kills the battery data on cdma
17:29.15Captnoordon wince it isn't called battery_id
17:29.17Captnoordits called
17:29.18CaptnoordDEX_Get_Batt_Slope_Kernel
17:29.26phhSlope ?
17:30.08phhCaptnoord: I think we killed something with  3D.
17:30.17phhdiamond ran every 3D app fine
17:30.20phhand now it no longer does
17:31.03phhdon't know what could be the reason though
17:31.05bzosome apps run and some don't?
17:31.14phhbzo: yeah
17:31.30phhI thaught blac's problems was only memory, but maybe it's not ...
17:32.18bzoit does seem weird that on the diamond, 3d is broken with more than 7mb
17:32.19Captnoordhmmmmm
17:32.29Captnoordbrb dinner
17:32.36Captnoordi'll check it when i'm back
17:35.57phhhum
17:36.00phhor maybe not
17:36.03phhthere is only one app that doesn't run
17:40.30phhhum neocore force close too
17:41.34bzoi get force close on neocore sometimes even before the 3d fix
17:42.27ToAsTcfh<4>[  603.687438] msm_cpufreq_target 550000 r 1 (245760-550000) selected 480000
17:50.05phhah
17:50.11phhneocore works again after reinstalling it.
17:50.36phhand kills android in the meantime ?!?
17:50.43phhah only mediaplayer
17:51.22GNUtooleviathan, what should I do? I spent the *whole day* making the driver compile and now it doesn't run:
17:51.34leviathanhmm
17:51.43GNUtooIt boots,I've the carrier logo,then it becomes black and then it reboots
17:51.59leviathanon which codebase do you work?
17:52.02leviathanwhich repo
17:52.07leviathanwhich branch
17:52.12leviathanwhich release?
17:52.15GNUtooyours + importing .31's msmfb driver
17:52.19leviathanhmm
17:52.35leviathanit seems that they have overwritten jumping adresses
17:52.38leviathanwhile in ours
17:52.46GNUtoojumping but where?
17:52.55leviathanin ours its practically vanilla from torvalds
17:52.58GNUtooand how could you know it ?
17:53.01GNUtooah ok
17:53.11*** join/#htc-linux Wr4i7h (~abc@89.180.151.129)
17:53.22leviathanthey have done a death sin in kernel hacking the android people
17:53.29leviathanput theire noses into base code
17:53.40GNUtooyes I modified memory
17:53.41leviathanwhere they dont have too do *anything*
17:53.50leviathannja
17:53.52GNUtoobut how can you know?
17:53.52leviathananyway
17:53.56leviathanuhm
17:54.06GNUtooyou've tried too?
17:54.12leviathanI tried recently to merge our kernel drivers into upstream again
17:54.14leviathanand yes
17:54.15leviathan:)
17:54.16leviathanboth
17:54.21GNUtooand?
17:54.25leviathanits pita
17:54.37leviathanthey hacked around in the basecode stuff of the kernel
17:54.45leviathanthats the reason why I'm only IMPORTING
17:54.52leviathanseparate functions
17:54.57leviathannot taking whole drivers
17:55.02GNUtooleviathan, do we have something like ramconsole?
17:55.14GNUtoobecause our bootloader doesn't support looking into the mem I bet
17:55.17leviathanits not runnable anymore with default linux kernel api
17:55.19leviathanI'm sorry
17:55.43GNUtoook so I'll let you import separate functions but I bet it won't work
17:55.52leviathanthere are a few things which have to be implemented totaly different if you use *real* linux
17:56.05GNUtoobecause one function depends on another which depends on another which depend on the modified memory
17:56.08leviathanmeans using the api not modding the api until it does what you want
17:56.31leviathanit will
17:56.52leviathanI didn't only make the xf86-video-msm work by importing headers
17:57.06leviathanI also extended our msmfb by some functions of codeaurora
17:57.10leviathanif you watch changelog :)
17:57.18GNUtoook
17:57.26leviathanwhat I need to do is to add the rest of the functions
17:57.31leviathanbut in order to do so
17:57.43leviathanI need to separate them from the rest of android stuff
17:57.52leviathanwhich they are reffering to atm
17:58.21GNUtoook
17:58.31GNUtootoo bad their new improved driver seemed nicer
17:59.41Captnoordre
18:00.04Captnoordmaybe we need to 0 it when we init it
18:00.11Captnoordand after that
18:00.14Captnoordwe can set it back
18:01.54GNUtooI want jtaaag kgggdb or at least serial
18:06.11phhGNUtoo: what about a good keyboard first ? :p
18:06.22Captnoordphh everything should be handled by libgl
18:06.27Captnoordregarding 3d
18:06.34Captnoordon official android devices
18:06.39GNUtoolol
18:06.41phhCaptnoord: not clock
18:06.45Captnoordhmmmm
18:07.02Captnoordwould libgl trigger our clock stuff correctly?
18:07.29phhCaptnoord: na but don't worry it works after all
18:07.35phhit was only basic android bug
18:07.40Captnoordk
18:07.46Captnoordokey
18:07.55Captnoordpreformance isn't different?
18:08.09phhit seems slower, but it's always hard to compare :/
18:08.15phhCaptnoord: see hw3d.c for clocks handling
18:08.17Captnoordnah its possible
18:08.19Captnoordk
18:08.22phhiirc clock is enabled on open
18:09.08leviathanGNUtoo: sry, was smoking
18:09.11GNUtoonp
18:09.12leviathanyou can have jtag
18:09.18leviathanhave fun with soldering ;)
18:09.21leviathaninside is a port
18:09.23leviathan:)
18:09.33GNUtooI started taking apart my headphones
18:09.38leviathanokee
18:09.40GNUtooI think I'll try serial first
18:09.46leviathanok
18:10.06leviathanbasically you can use every htcdatacable
18:10.15GNUtooah?
18:10.20GNUtooI've a headphones cable
18:10.44leviathanand there is a webstore for ordering the conncetors+platine all in one
18:10.50phhGNUtoo: uart3dm is on usb D+/D- pins
18:11.17GNUtoophh, thanks , is there a howto somewhere, which kind of max do I need? 3232 is ok?
18:11.44*** join/#htc-linux MrPippy (~pip@adsl-75-42-163-160.dsl.sndg02.sbcglobal.net)
18:12.08phhGNUtoo: hum, don't know
18:12.31phhhi MrPippy
18:12.37MrPippyhey
18:12.47phhor back, but I think you just woke up ?
18:12.55phh(stupid timezones.)
18:13.22ToAsTcfhphh: do u have this when oc? /sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed
18:13.25*** join/#htc-linux marcin_ (~marcin@chello089078146089.chello.pl)
18:14.09phhToAsTcfh: iirc it depends on the governor
18:14.41phhToAsTcfh: I have it but it's marked as "unsupported"
18:14.42ToAsTcfhadb shell cat  /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor
18:14.42ToAsTcfhondemand
18:14.43Captnoordwhats HW3D_NUM_REGIONS
18:14.46CaptnoordI mean
18:14.48Captnoordwhats the number
18:14.51Captnoordgrps
18:14.52Captnoordgreps
18:14.54phhCaptnoord: 3
18:15.03phhToAsTcfh: then you can't use setspeed
18:15.08phhyou must use userspace to use setspeed
18:15.16CaptnoordI think thats imem_config
18:15.21CaptnoordI guess
18:15.23ToAsTcfhsetspeed?
18:15.49CaptnoordI can't find it
18:15.50Captnoord:S
18:15.51ToAsTcfhcat  /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_governors
18:15.51ToAsTcfhmsm7k userspace ondemand performance
18:16.12ToAsTcfhcat  /sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed
18:16.12ToAsTcfh<unsupported>
18:16.12ToAsTcfh<PROTECTED>
18:16.59phhCaptnoord: ?!?
18:17.11phhCaptnoord: no num_regions has nothing to do with that
18:17.23phhCaptnoord: when you mmap /dev/hw3d, you get 3 regions:
18:17.25phh"pmem" in smi
18:17.28phh"pmem" in ebi
18:17.42phhand the 0x800000xxxx regs which drives the 3D engine
18:18.02ToAsTcfhphh: if i echo to scaling_cur_freq nothing changes in it :(
18:18.04phhthat's what HW3D_NUM_REGIONS=3 means.
18:18.26phhToAsTcfh: echo userspace >  /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor
18:19.41ToAsTcfhit wont let me use the > comand
18:19.45ToAsTcfh:(
18:19.49CaptnoordI know
18:19.50phhlol?
18:20.08ToAsTcfhadb shell echo userspace  /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor
18:20.09ToAsTcfhuserspace /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor
18:20.19phhToAsTcfh: why don't you press the > ?
18:20.28ToAsTcfhadb shell echo userspace >  /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor
18:20.28ToAsTcfhbash: /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor: No such file or directory
18:20.36phhescape the >
18:20.41phhadbd shell echo userspace \> blabla
18:20.45phhadbù
18:20.46phh*
18:20.47phhbah
18:20.47phh.
18:21.59CaptnoordI think why its slower is because the pll's aren't changed when enabling 3d
18:22.00ToAsTcfhsweet it worked ;P
18:22.11Captnoordaren't increased or what ever
18:22.24Captnoordlike we you can see in the 7x30 clock file
18:22.31Captnoord-we
18:26.44ToAsTcfhphh: still oesnt work :(
18:26.50phhprecise doesn't work.
18:27.16*** join/#htc-linux Reefermattness (~MattGNM@c-76-18-79-170.hsd1.nm.comcast.net)
18:27.21phhCaptnoord: grp clocks derivates from which clock ?
18:27.25phhwhich pll i mean
18:27.49ToAsTcfhmeanin i can change it to userspace but i still cant echo to set speed or anything else and it stick
18:28.07phhToAsTcfh: precise "can't echo"
18:28.54Captnoordgood question
18:28.57Captnoordlemme see
18:29.14phhCaptnoord: I'd love overclocking grp.
18:29.19Captnoordhehe
18:29.34phhhaving same perfs as android devices, but on a VGA screen :D
18:29.53phhwell I don't think we can overclock it that much.
18:29.56ToAsTcfh<PROTECTED>
18:30.09phhToAsTcfh: what do you mean "you're suck on pll1" ?
18:30.13phhan 528 is on pll2 iirc.
18:30.30CaptnoordI think axi
18:30.33ToAsTcfhi wont go any higher then 480
18:31.21phhhow do you check current speed ?
18:32.14ToAsTcfhcat /sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq
18:32.20CaptnoordI think pll 3
18:32.38*** join/#htc-linux Reefermattness_ (~MattGNM@c-76-18-79-170.hsd1.nm.comcast.net)
18:32.44Captnoorddo we have one a 3?
18:32.45Captnoord:P
18:32.46Captnoordlol
18:32.55Captnoordi'm reading other shit of course
18:33.43*** part/#htc-linux Phil_123 (~anonymous@221.127.43.91)
18:34.05Captnoordcr2 knows
18:34.52phhI need a proper way of handling speaker on/off groumpf
18:36.06ToAsTcfhphh: if i compile with acpuclock.c set at 528000 and i set the config to the same it will have no issues. but as soon as i in crease the clock speeds it either wont boot(if the config is higher then the acpuclock speed) or will boot but wont go over 480 (when the clock in acpuclock is higher then the configs clocck)
18:36.26*** join/#htc-linux Wr4i7h_ (~abc@89.181.81.56)
18:36.29phhToAsTcfh: perflock is disactivated ?
18:36.34ToAsTcfhyes
18:36.53phhwhat's in cpuinfo_max_freq ?
18:37.20ToAsTcfhright now it 550000
18:37.37ToAsTcfhwhat i set the config for
18:37.52ToAsTcfhbut acpuclock.c is at 610000
18:37.58ToAsTcfhso itll boot
18:38.24ToAsTcfh<4>[  648.107635] msm_cpufreq_target 550000 r 1 (245760-550000) selected 480000
18:38.34phhcat cpuinfo_max_freq > scaling_min_freq
18:39.32ToAsTcfh# cd /sys/devices/system/cpu/cpu0/cpufreq
18:39.32ToAsTcfh# cat cpuinfo_max_freq > scaling_min_freq
18:39.33ToAsTcfhwrite: Invalid argument
18:39.52*** join/#htc-linux k_linux (~IceChat7@41.248.145.180)
18:40.10phhToAsTcfh: funny.
18:40.14phhk_linux: I hate your damn connection :p
18:40.21MrPippyoh damn...the one part i still have to figure out for wifi is getting the nvram from nand
18:40.23k_linuxphh: me too :D
18:40.27phhMrPippy: ?
18:40.38phhMrPippy: like copy pasting the code from TI's loadit ? :p
18:40.47MarkinusMrPippy: look in your /windows directory, nvram.txt
18:40.50k_linuxhttp://www.4shared.com/dir/32335779/5c5ec6f1/sharing.html
18:41.06MrPippyyeah i've done it, just have to figure out where to put the script
18:41.15phhMrPippy: ah.
18:41.16k_linuxphh: as it was too slow I uploaded the files one by one
18:41.19phhMrPippy: init ?
18:41.22Captnoordstatic struct clk_freq_tbl clk_tbl_grp[] = {
18:41.22Captnoord<PROTECTED>
18:41.23Captnoord<PROTECTED>
18:41.23Captnoord<PROTECTED>
18:41.23Captnoord<PROTECTED>
18:41.23Captnoord<PROTECTED>
18:41.23Captnoord<PROTECTED>
18:41.24Captnoord<PROTECTED>
18:41.24Captnoord<PROTECTED>
18:41.48phhCaptnoord: yes but no.
18:41.52Captnoordyes but no
18:41.52Captnoordyea
18:41.53MrPippyyeah i think it'll have to be init, since i don't run loadit or need any other wlan_loader
18:41.56phhI think we have no pll3 actually :p
18:41.56Captnoordpll0 can also be used
18:42.01Captnoord?
18:42.16CaptnoordI dono
18:42.26phhah no sorry we have a pll3.
18:42.34phhMrPippy: and init is fine
18:42.37phhMrPippy: move it for TI too
18:42.48k_linuxphh: enjoy !! :D
18:42.50phhno need to do that thing at every wifi load
18:42.56Captnoordthe 7500 stuff talks about the presents of pll3
18:42.57Captnoordbut
18:43.01CaptnoordI actualy don't know
18:43.03Captnoord:S
18:43.12phhCaptnoord: just read the "clk_src" bits
18:43.21Captnoordk
18:43.41phhk_linux: so everything works but BT ?
18:43.43ToAsTcfhphh: found this in the boardheroc.c > http://en.pastebin.ca/1813961
18:43.50MrPippyits not a problem to do it at every load, if you already have the calibration file then it doesn't do anything
18:43.53k_linuxphh: yes
18:44.09phhk_linux: sounds hard to believe :p
18:44.14phhMrPippy: yes right.
18:44.22k_linuxphh: the Gsensor work on abduction but not in setting
18:44.34phhToAsTcfh: that's only perflock thing
18:44.37k_linuxphh: thx to libhtc_acoustic.so
18:44.39k_linux;)
18:44.52phhk_linux: uh ?
18:44.58phhoh you had sound problems because of it ?
18:45.04phhwe have to make it work, and even recode it.
18:45.26k_linuxphh: no no it's fine I misunderstund u
18:45.38phhok
18:45.40MarkinusMrPippy: On leo they loading the firmware and the nvram on boot, and then disabling wlan if not enabled. The fiormware loading is only on boot
18:45.42ToAsTcfhToAsTcfh gets a gun
18:46.11MrPippywe can move it for TI later if you want, this commit is big enough already
18:46.22phhMrPippy: lol as you want
18:46.25k_linuxphh: going back to linux, see ya
18:46.29MrPippythe module loads its own firmware and nvram
18:46.32phhI don't really care, I just think it's better that way :p
18:47.08phhthe only things I care about, is having the same build for everyone, and code having good indentation :D
18:48.04MarkinusMrPippy: yes, I know. Leo has (alwost) the same chip. I woanted to say they didn'T do it on every wlan enabling, only once at boot . .
18:48.37MrPippythey load the module at boot?
18:50.00MarkinusMrPippy: Yes, I saw it an dmesg
18:50.00ToAsTcfhphh: http://en.pastebin.ca/1813987 is all i can think is the issue. im out of ideas :(
18:50.08MarkinusI disabled wlan and they laoding it on boot
18:50.13Markinusand then disabling this
18:51.11MrPippyhmm android userland would need (even more) work to make sense of the module loaded all the time
18:51.28MrPippyi haven't looked real closely, but i think devour loads the module at wifi poweron like usual
18:52.08MarkinusMrPippy: I can only say what I saw . . maybe they have a good PM in the chip and they didn'T need this . .
18:52.45bzoMrPippy: I saw that you had a fix for diam500 bluetooth back in december. Was bt working with donut back then?
18:53.06MarkinusMrPippy:  I thing this is one reason on the other chip to unload .. to deactivate and safe power . .
18:53.30MrPippyi have read it has good pm, and in wince very little happens when it gets turned on/off--just one gpio, no vregs
18:54.55MarkinusMrPippy: maybe you can do it in this way .. . enabling/disabling of wlan could be faster
18:55.49MrPippyi've never tested BT (or at least never seen it working), that fix was mostly because the gsm bt gpios were stomping on something (usb i think)
18:56.05MrPippyi did put in the BT gpios that i saw in wince, but never tested BT out
18:56.52MrPippywifi starts in ~5 seconds, i'm happy with that :-)
18:57.48GNUtoohow does it differenciate between audio and serial? you've got serial out of the box and should short 2 pins for audio?
18:58.13*** join/#htc-linux k_linux (~klinux@41.248.145.180)
18:58.52bzoMrPippy: ok, thx. Perhaps some more init needs to be done with bt. With stinebd's fix for bt in the rootfs, it is not working for me in eclair
18:59.24MarkinusMrPippy: leo need 2 :) But I think the most important this is, that you made it work! :)
19:00.57*** join/#htc-linux M1DLGpc (~M1DLGpc@bb-87-81-252-83.ukonline.co.uk)
19:01.09MrPippyis android running at all on leo, or just linux right now? and do they have a git somewhere?
19:06.25MarkinusMrPippy: no, no linux . . it's very early state
19:06.32Markinuswe have no clean boot to initrd
19:07.07MrPippyhmm thats what i thought
19:07.07Markinus(what I explained was wince dmesg)
19:08.32Markinusbut you can look no nexus .. shoulb be the same chip . .
19:09.40phhMrPippy: to have module always loaded with android userspace, you can do a fake wlan.ko module
19:09.50phhbut I don't think this is needed
19:10.24MrPippyyeah
19:10.48phhiirc they load the module when needed
19:10.50*** join/#htc-linux ALoGeNo (~alogeno@unaffiliated/alogeno)
19:11.14MrPippyso how about mapping poweroff to reset? if poweroff just hangs on every device, then no downside
19:11.25ToAsTcfhphh: do u know how i would add the clock speeds to my init.rc? like u guys do with ur default.txt.
19:11.25phhoh you haven't done it yet ?
19:11.52phhToAsTcfh: not in init.rc
19:11.53MrPippyno i'll try it out now, push it later if it seems good
19:12.27phhok
19:14.12k_linuxphh: did u download it ?
19:14.20phhk_linux: yup
19:14.51k_linuxphh: so is there something to do to have BT ?
19:15.20phhhum wait
19:15.25k_linuxok
19:15.33phhI want to do the test by myself but I can't right now
19:15.52k_linuxphh: np
19:16.25phh(well, i could, but I don't feel like leaving my bed :p)
19:16.40k_linux:D
19:16.42ToAsTcfhlazy ass
19:16.47ToAsTcfh:P
19:16.58ToAsTcfhdid my config look ok?
19:17.06phhToAsTcfh: don't know :/
19:17.11MrPippyyay putting the battery cover back on my rhod
19:17.16ToAsTcfhdamn
19:18.15phhMrPippy: :))))
19:19.01*** join/#htc-linux Wout (~wout85@s55927d21.adsl.wanadoo.nl)
19:19.05Woutwhoops, wrong button
19:19.52phhk_linux: oh right, it needs several changes actually
19:20.01phhso it will wait a bit
19:20.48phhit's not hard, but there are at least 4 or 5 steps to do
19:20.56phhand I'm not really sure of any of them.
19:21.08*** join/#htc-linux M1DLGpc (~M1DLGpc@bb-87-81-252-83.ukonline.co.uk)
19:21.52k_linuxphh: ok will wait then
19:22.46MrPippyok pushed the reset patch
19:23.13k_linuxphh: I'll share it on xda so everyone enjoy it ! ;)
19:23.27WoutMrPippy: how's the wifi commit coming along?
19:23.28phhyou mean before or after I fix BT ?
19:24.06k_linuxphh: before but if u want it will wait
19:24.06moishaSPB<MrPippy> poweroff works fine on Kaiser. where u pushing to ?
19:24.22phhmoishaSPB: we're speaking about a totally different kernel
19:24.32moishaSPBk
19:24.33phhfor "recent" devices
19:24.56MrPippykernel stuff is almost ready to go, rootfs needs one more piece (nvram)
19:24.56Woutwait? BT also almost working? So BT+Wifi+backlight+gsensor+pm in just two weeks? :D you guys rock
19:25.07phhWout: na
19:25.10phhnot for you the BT
19:25.22phhwell
19:25.24phhBT isn't hard
19:25.26GNUtoohi,where could I find help for identifying the pins for making a cable?
19:25.27phhit's just annoying
19:25.55*** join/#htc-linux cr2_ (~cr2@ip-109-85-52-191.web.vodafone.de)
19:26.53MrPippyi can't figure out why the wifi drivers (both ti and bcm) statically allocate these buffers in the kernel
19:26.55cr2_NetRipper: ping
19:27.25*** join/#htc-linux cr2_ (~cr2@ip-109-85-52-191.web.vodafone.de)
19:27.29phhMrPippy: because they are wince's driver.
19:27.52phhGNUtoo: http://forum.xda-developers.com/showthread.php?t=595791 ?
19:28.03GNUtoothanks a lot I'll look
19:28.12phhk_linux: GROUMPF
19:28.18phhyou could have put updated gsensor.
19:28.26cr2_phh: i think i've found the problem on leo
19:28.29MrPippyi think the android guys actually had to add it to the bcm driver though
19:29.13k_linuxphh: well I didn't try a new eclair build so I used the which I have
19:29.21cr2_phh: the evil androids could have been more friendly to us
19:29.22phhNB: having two android devices is no good idea.
19:29.32phhespecially when doing commands like adb shell rm /sdcard/data.img
19:29.37phhcr2_: why ?
19:30.04cr2_phh: the smd id list in smd_private.h is fucked again
19:30.10phharf
19:30.16phhyes that's unfair :/
19:30.20cr2_phh: they could have written "this is for amss $foo"
19:30.49cr2_<PROTECTED>
19:30.59cr2_so we are shifted by 1
19:31.28cr2_after the first entry, and the irq handler picks some junk from smem
19:33.07phhcr2_: ha that's all your irq problem ?
19:34.21cr2_phh: NetRipper's
19:34.27phhright.
19:34.37GNUtoophh, thanks a lot I didn't thought my multimether could probe the pins becuase they are tiny but it can
19:34.52cr2_lol
19:39.44phhk_linux: everything force closes -_-'
19:40.11phhacore, ime, media
19:40.18k_linuxphh: you are using it on diam that's why
19:40.22phhno.
19:40.28phhrhod
19:41.02k_linuxphh: well reboot it it should work
19:41.10phhhum ok
19:42.12*** join/#htc-linux M1DLGpc (~M1DLGpc@bb-87-81-252-83.ukonline.co.uk)
19:44.55*** join/#htc-linux JesusFreak316 (~JesusFrea@pool-173-65-77-190.tampfl.fios.verizon.net)
19:46.45*** join/#htc-linux [acl] (~abel@cpe-69-203-142-192.si.res.rr.com)
19:47.34phhk_linux: seems to be wrong data.img
19:47.46k_linuxphh: what do u mean ?
19:47.52*** join/#htc-linux houmles (~houmles@vodnik.dkm.cz)
19:47.52phhmy crashes
19:48.13k_linuxphh: u tried with eclair data.img ??
19:48.20phhkind of.
19:48.39phhI removed data.img, then I started with wrong system.sqsh then with correct one
19:48.55k_linuxphh: ok hope it'll work now
19:49.27k_linuxphh: u know the 1st boot is so slow
19:49.35phhyeah :/
19:49.54phhand it's not even to compile the jar, it's just to make obex blabla
19:49.54phhpfff
19:49.56JesusFreak316Hey, I tried repackaging my system.sqsh as a system.ext2 for write access to the system folder and it boots but none of the buttons work; they turn the screen black and returns to the home screen, making it unusable. What would cause that?
19:50.42phhk_linux: do you know how to mmutrace i2c in haret ?
19:50.57k_linuxphh: no
19:51.14k_linuxbut I can learn :)
19:51.44phhnot today.
19:51.49k_linuxphh: as I did to have 2.1 working
19:52.10k_linuxphh: will wait then
19:52.57k_linuxphh: is it possible to use the latest htc desire rom on our devices? u know that's allready wvga
19:53.16phhk_linux: I don't think so, different CPU
19:54.01*** part/#htc-linux JesusFreak316 (~JesusFrea@pool-173-65-77-190.tampfl.fios.verizon.net)
19:54.03k_linuxphh: we can't use libs or apps at least ??
19:54.03*** join/#htc-linux JesusFreak316 (~JesusFrea@pool-173-65-77-190.tampfl.fios.verizon.net)
19:54.08*** part/#htc-linux JesusFreak316 (~JesusFrea@pool-173-65-77-190.tampfl.fios.verizon.net)
19:54.11phhnot libs
19:54.14*** join/#htc-linux JesusFreak316 (~JesusFrea@pool-173-65-77-190.tampfl.fios.verizon.net)
19:54.18*** part/#htc-linux JesusFreak316 (~JesusFrea@pool-173-65-77-190.tampfl.fios.verizon.net)
19:54.18phhand only non native apps
19:54.28k_linuxphh: framework maybe
19:54.33phhmaybe
19:54.59[acl]sounds like you guys need to try it out.
19:54.59phhbut I think they use some native libs
19:55.33*** join/#htc-linux JesusFreak316 (~JesusFrea@pool-173-65-77-190.tampfl.fios.verizon.net)
19:55.36k_linuxphh: I think so too
19:56.05phhk_linux: now IME won't stop crashing -_-'
19:56.19*** join/#htc-linux M1DLGpc (~M1DLGpc@bb-87-81-252-83.ukonline.co.uk)
19:56.41k_linuxphh: change it with another it will be fine
19:56.55phhthere is one included ?
19:57.09k_linuxphh: wait a minute
19:57.39cr2_phh: amss progress: SMEM_USABLE_RAM_PARTITION_TABLE, SMEM_POWER_ON_STATUS_INFO
19:57.42*** join/#htc-linux swc|666 (~infidel20@unaffiliated/swc666/x-4934821)
19:57.51phhcr2_: for leo ?
19:57.58cr2_yes
19:58.21phhpff
19:58.28phhnothing for good old devices ? :'(
19:58.30cr2_phh: so the memory bank data can be picked from SMEM_USABLE_RAM_PARTITION_TABLE
19:58.51k_linuxphh: no, u have to adb install it :S
19:59.28phhk_linux: ok
19:59.49k_linuxphh: u know it's still experimental
19:59.57phhnoway !
20:00.07k_linux?
20:00.09phhbut it works for you ?
20:00.10ToAsTcfhphh: cant i do it like this? CONFIG_CMDLINE="mem=64M console=ttyMSM,115200n8,acpuclock.oc_freq_khz=550000"
20:00.15phhToAsTcfh: sure
20:00.20phhToAsTcfh: not with a ,
20:00.23phhbut with a space.
20:00.25k_linuxphh: yes ofcourse
20:00.32phhk_linux: weird
20:00.37phhit complains about a missing native lib
20:00.55k_linuxphh: which one ?
20:01.06phhnot sure
20:01.18*** join/#htc-linux pH5 (~ph5@e178198007.adsl.alicedsl.de)
20:01.21ToAsTcfhphh: like this? CONFIG_CMDLINE="mem=64M console=ttyMSM,115200n8 acpuclock.oc_freq_khz=550000"
20:01.23phhW/dalvikvm( 1112): No implementation found for native Lcom/android/inputmethod/latin/BinaryDictionary;.openNative (Landroid/content/res/AssetManager;Ljava/lang/String;II)I
20:01.25phhToAsTcfh: yes
20:01.47k_linuxphh: will add it
20:01.53phhk_linux: and 3G connection doesn't work -_-'
20:01.57phhI feel really unlucky today.
20:02.04k_linux:)
20:02.17phhand again it doesn't seem to be your fault, ril is fine
20:02.19phhpppd too
20:03.02*** join/#htc-linux m3dlg (~m3dlg@bb-87-81-252-83.ukonline.co.uk)
20:03.04phhk_linux: hum, htc sense keeps crashing
20:03.30phhE/AndroidRuntime( 1392): java.lang.RuntimeException: Unable to start activity ComponentInfo{com.htc.launcher/com.htc.launcher.Launcher}: java.lang.IllegalStateException: The wallpaper must be a BitmapDrawable.
20:03.33phhthat's damn explicit.
20:03.45k_linuxphh: push the power button and select home as default
20:03.55phhdidn't you say sense works ?
20:04.29[acl]what device are you guys testing sense on?
20:04.31k_linuxphh: yes but the one I shared doesn't have all htc's apps
20:04.41phhk_linux: pfff
20:05.46k_linuxphh: well u can try the original one it's linked here http://forum.xda-developers.com/showthread.php?t=613826
20:07.50k_linuxphh: I replaced libhtc_acoustic and added eclairhero.build, sensors.default.so
20:09.01cr2_phh: 00000001 0002c868 00000010 00000000  SMEM_BATT_INFO
20:09.15phhcr2_: you have a smem batt info ?
20:09.22phhyou're damn lucky.
20:09.23cr2_on leo
20:09.35k_linuxacl: blac and rhod
20:10.07phhk_linux: and diam
20:10.31phhfor bt I have to test it on diam annyway
20:10.34k_linuxphh: yes forget it
20:10.53phhevery htc devices that are >=VGA :p
20:11.42*** join/#htc-linux GlemSom (~glemsom@0x5da34bca.cpe.ge-1-1-0-1105.sdnqu1.customer.tele.dk)
20:11.59phhGlemSom: !
20:12.03phhhi
20:12.11GlemSomphh, Hi :)
20:12.15phhthere is a patch for the tiwlan driver
20:12.24makkonenthere is?
20:12.30ToAsTcfhok so i can rule that out as the issue
20:13.02phhGlemSom: HTC guys commited a patch marked as "Do not merge" :-)
20:13.04GlemSomphh, Maybe we should have kept tiwlan in a git repo then? :P
20:13.12GlemSomphh, what, really? :P
20:13.44phhhttp://android.git.kernel.org/?p=platform/system/wlan/ti.git;a=commitdiff;h=849c6d4186a7fcd4d25bcff6cc2e6bc25f8c4696
20:14.28phhjust reverse it
20:14.41phhif you really want a git repo I can do it, but I really don't want.
20:15.23*** join/#htc-linux leaigor (~laigor@188.134.16.241)
20:15.35GlemSomphh, Well, sofar this has been the only change we needed for the tiwlan driver... Så guess it's not worth the trouble yeat... :)
20:16.08phhI guess we should have one to backport google's changes though
20:17.15makkonenwhy not just build off the repo from android.git.kernel.org? are any local changes being made to it?
20:18.00phhmakkonen: and then they break voluntarly their tree, and we have to change it again.
20:18.09phhwell, not neceserraly voluntarly.
20:18.21makkonenjust... inconsiderately.
20:18.41*** join/#htc-linux JesusFreak316 (~JesusFrea@pool-173-65-77-190.tampfl.fios.verizon.net)
20:24.04WoutMrPippy: just saw your commit, do I just need to download the latest zimage/modules for wifi to work? (there's already a build up at glemsons site)
20:24.17phhhe commited .. ?
20:24.41Woutyeah, it says 'Enable wifi on rhod'
20:24.44[acl]you probably need the rootfs changes too. im not sure he is done with that
20:24.56phhputs his glasses on
20:24.58phhWout: ok where ?
20:25.03Wouthttp://gitorious.org/linux-on-qualcomm-s-msm/linux-msm
20:25.19phhah this one.
20:25.25phhyou need the updated rootfs
20:25.41makkonenwhich doesn't exist yet
20:25.56phh:p
20:26.00k_linuxphh: so does it work at last??
20:26.05Woutok
20:26.08phhk_linux: kind of
20:26.16Woutneeds to learn to be patient
20:26.37GlemSomohh, it's compiling atm... will patch after it's done with this compile...
20:26.45phhGlemSom: hehe
20:26.57k_linuxphh: well it's better than nothing ;)
20:27.07phhk_linux: but I want htc sense -_-'
20:27.32phhOUCH
20:27.33phhdonut init
20:27.33k_linuxphh: if u r very patient I can share it :D
20:27.54Unholyhey phh did you check out the quake 3 port for android
20:28.12Unholyit works but slow , i imagine cus of the vga screen
20:28.15phhUnholy: how ? it needs a keyboard, and I have no 3D on rhod.
20:28.24Unholydont you have a raph?
20:28.25GlemSomphh, We agree it's this right -> http://pastebin.com/eLhBTr9z
20:28.27phhUnholy: diam
20:28.39phhGlemSom: *reverse* it
20:28.43Unholydoes it ahve hw3d?
20:28.46Unholyhave*
20:28.48phhUnholy: yes
20:28.50GlemSomphh, Yeah - i did
20:28.54phhGlemSom: ok
20:28.56phhthen yes it's it
20:29.02Unholytry it then Xd
20:29.11phhUnholy: anyway, it's slow on all msm7k devices
20:29.18Unholyhttp://code.google.com/p/kwaak3/
20:29.38phhafaik.
20:30.32Unholywhats the clock of the hero? 528?
20:30.38phhyes
20:30.46phhmaybe not stock mmmm
20:31.04Unholyi see well the guy says it rins at 14fps no sound
20:31.08Unholywith sound 11
20:31.15phhnot about q3
20:31.20Unholybut thats a hvga scrteen
20:31.31phhah right.
20:31.38phhanyway 14fps is unplayable -_-'
20:31.46phhI really need to add a scaling option.
20:31.54Unholylower graphics gives you 5 fps more
20:32.02Unholylowering*
20:32.08Unholythe gy even got networking going
20:32.11Unholyguy*
20:32.36Markinuscr2_: I made the chhanges for MDP/LCDC . . we need the 16 Bit color stuff and I didn't know what the bot   #define LCD_MD_VAL_40MHZ 0x0005FFCF
20:32.38Markinus#define LCD_NS_VAL_40MHZ 0xFFD41B49
20:32.38Markinus# defines are
20:32.42Unholyi lowerd evrything within the game but it runs the same
20:32.44phhgimme a usb slave <-> ethernet convertor and I'm happy Unholy :p
20:32.48phhbut I have some doubts.
20:33.04*** join/#htc-linux andycandet (~andycande@188.26.135.174)
20:33.12Unholyhe succeded in playing against him self over the internet
20:33.14Unholywith hes pc
20:33.28phhsome people are crazy.
20:33.40Unholybut he had to change something dont remeber what it was
20:33.42UnholyXD
20:33.51polyrhythmiclol
20:33.56phh(noone to say I'm even crazier ? thanks)
20:34.10*** part/#htc-linux k_linux (~klinux@41.248.145.180)
20:34.13polyrhythmicyou're crazy as you wanna be phh
20:34.17polyrhythmicI'm not gonna stop you :)
20:34.40*** join/#htc-linux andycandet (~andycande@188.26.135.174)
20:34.42makkonenthat's more sane than the guy who runs 40-man raids on WoW where he controls all 40-men.
20:34.58Unholywasent that proven to be fake?0o
20:35.14phhit was only 25.
20:35.19phhmakkonen: got anything with battery smem ?
20:35.41polyrhythmicI have a Polaris user trying to boot android who continually gets 'Error waiting for notbusy' during boot
20:35.55polyrhythmicI found the kernel source for the error but I don't fully understand it... can anyone explain it?
20:36.12phhpolyrhythmic: I2C error ?
20:36.32makkonenphh: no. got sidetracked with apn junk to send to babijoee. I'm not really sure what I should be trying to track down. nothing jumped out at me from the heroc source.
20:36.49polyrhythmicyes phh I found the i2c code, but what causes i2c errors when booting from haret?
20:37.22phhmakkonen: pd 0x01f00000+0xfc140+18*4 4
20:37.31phhpolyrhythmic: can be anything
20:37.42phhblac users also have this problem
20:37.44phhit's really annoying
20:37.58polyrhythmichmmm, good to know
20:38.51polyrhythmicmust be irritating
20:39.06phhit just kills their wakeup.
20:39.11andycandethello all
20:39.13*** join/#htc-linux cr2_ (~cr2@ip-109-85-52-191.web.vodafone.de)
20:39.18andycandeti got a problem with android here
20:39.28polyrhythmicI got a couple i2c errors on my recent boot, maybe there's some solar flares going on :P
20:39.32andycandetdownloading apps from market fails
20:39.43phhandycandet: yeah market is kind of down
20:39.57andycandetoh, it was the market
20:40.12phhandycandet: not if you're speaking about last babijoee's build
20:40.13makkonenis it the market or is it the new xdandroid?
20:40.31[acl]im in pain right now looking at this old android boot screen .. :-(
20:40.33andycandetyes, i got the latest xdandroid (27/02)
20:40.40phh[acl]: lol
20:40.44phhandycandet: yes right this one is wrong
20:40.45andycandetwith the latest kernel
20:41.04makkonenyeah, the newest xdandroid fails to download from the market.
20:41.13andycandetok... thanks
20:41.18*** join/#htc-linux k_linux (~klinux@41.248.145.180)
20:41.53polyrhythmic[acl]: just replace your /data/media/bootanimation.zip and /system/bin/bootanimation with a different boot
20:42.00GlemSomphh, New build with tiwlan changes ready.
20:42.10phhGlemSom: great
20:42.21phhpolyrhythmic: bootanimation binary also needs to be changed ?!?
20:44.20GlemSomphh, I wonder why they decided to revert the patch in the first place... Does it break anything else in the tiwlan driver?
20:44.32[acl]once you cancel the 1001 windows that pop up.. not so bad. i can really butcher up this image and remove the crashing apps.
20:45.10phh[acl]: which one are you speaking about ?
20:45.36[acl]the 2.1 sense you guys were talking about earlier.
20:45.48phhah right
20:46.17phhimho it sucks, but it can have full BT working.
20:46.42phhwell it doesn't suck, there is just no improvement
20:48.08polyrhythmicphh: AFAIK, yes....
20:49.08polyrhythmicphh: I'm not sure if all binaries work for QVGA.  Actually, I wish I knew how to make a bootanimation QVGA-compatible, I google and can't find any answers
20:49.23Unholybtw someone should tell babyjoe that changing lcd.density to 240 makes some games and apps probly all mest up, so 210 is still the best
20:49.31cr2_phh: is it possible to parse SMEM_LOG_EVENTS ?
20:49.57GlemSomI just saw this news a few weeks ago... sounds quite interresting imo -> http://www.myriadgroup.com/Media-Centre/News/MYRIAD-BRINGS-3x-FASTER-APPLICATIONS-RICHER-GAME-GRAPHICS-AND-BETTER-BATTERY-LIFE-TO-ANDROID.aspx
20:49.58phhcr2_: don't know
20:50.06makkonenphh:  pd 0x01f00000+0xfc140+18*4 4? What's supposed to be there? cuz it's just 0s. I'm not sure what I should be looking for.
20:50.08cr2_ok
20:50.25phhmakkonen: they are just 0s then :/
20:50.45makkonenis something supposed to be there?
20:50.47cr2_makkonen: pd 0x01f00000+0xfc140 are battery dex data on cdma raph/diam
20:50.51phhGlemSom: that has already ran all over xda :p
20:51.04cr2_makkonen: it's documented in wiki
20:51.04phhmakkonen: if it was like hero, battery percentage
20:51.09makkonenah
20:51.11GlemSomphh, Ohh :/   I'm soooo behind here! :P
20:51.15*** join/#htc-linux M1DLGpc (~M1DLGpc@bb-87-81-252-83.ukonline.co.uk)
20:51.16phhcr2_: is there anything after the first 5 words ?
20:51.29cr2_phh: on cdma ? don't think so
20:51.41phhcr2_: I noticed.
20:51.47phhGlemSom: we haven't implemented it yet though
20:52.06GlemSomphh, one step at the time.. right :P
20:52.44phhand I don't really fell like playing with that :/
20:52.52makkonenphh: there's something at 0xfc194, but that's it.
20:53.01MrPippythe fc140 area is also used for cdma statistics (signal strength, etc.) and the values get switched by sending a dex call
20:53.14phhmakkonen: ok, that's all then.
20:53.18MrPippyiirc whenever i tried to dump it, the cdma stuff would show up and not the battery stats
20:53.30makkonenthat's... complicated.
20:53.34MrPippyif you want to see the battery stats mmutrace is the way
20:55.07phhMrPippy: or create a haret cmd to call the dex :p
20:55.54adamwMrPippy: awesome job on wifi - didn't test it yet but I trust you :P
20:56.24MrPippyhehe thanks it's not done yet, still have to push the rootfs and then we can test
20:56.44[acl]can you at least share the changes so i can replicate it on my own root fs :-)
20:57.14Woutsee you guys later
20:57.16Woutsocial time
20:57.30adamwMrPippy: next job - sound ;)
20:58.01phh(I already started the job about the speaker btw)
20:58.02Woutif you guys keep up this pace, there wont be much left to fix :P
20:58.03adamwphh: you say bt isn't hard but it's annoying - is it sufficiently not-hard that I could help with it? is it just some kind of dull trial-and-error thing?
20:58.11Markinusadamw: There we have to hope that cr2_ maybe have a bit time to help ..
20:58.24adamwphh: awesome
20:58.27phhadamw: the only way is to port a devour or galaxy ROM.
20:58.33makkonenadamw: I LOVE dull trial and error things.
20:58.37phhadamw: I don't mean software sound thouogh.
20:58.39adamwoh, kay. a bit too tricky for me I guess :/
20:58.48adamwmakkonen: yeah i don't mind doing them either, which is why I asked
20:58.55phhI mean just the electrical part
20:59.00phhie you'll be able to make calls on it but that's all :p
20:59.01MrPippynexus one bt should be the same
20:59.02adamwphh: anything helps
20:59.11phhthe DSP part is really ....
20:59.14phhI don't know
20:59.22phhit just don't work and no clue why
20:59.44adamwowch
21:00.23adamwhas to admit latest energyrom is so good it makes it hard to go for android, heh
21:03.17*** join/#htc-linux Zeman4323 (~zemanm@65.30.176.154)
21:03.31makkonenmrpippy: so what do I need to do to mmutrace the battery areas?
21:04.29MrPippyaddlist mmutrace 0xbc0fc140 0x10 should do it
21:04.42*** part/#htc-linux k_linux (~klinux@41.248.145.180)
21:04.56IceBoneWhy has my android broken down?
21:05.03makkonenok. couldn't figure out the virtual address.
21:05.09IceBoneDuring boot it says: Failed to find system.img ong SD Card.
21:05.14IceBoneon*
21:07.11MrPippyyeah 'dump mmu 2' is the way to find all the virtual mappings but each driver could use different ones, for cdma 0xbc covers all of smem and i think most drivers use that one
21:07.36makkonenthanks
21:08.04makkonenthere's some data. neat.
21:10.08makkonenstill nothing from fc188 though
21:10.30phhmakkonen: then you don't have it, no need to search further
21:10.56makkonenyup
21:11.32MrPippythe only battery-related value not around fc140 is vref at fc0e0, for me its constant at 0x4d5 (1237) and i can't figure out what its for
21:12.47phhMrPippy: maybe vbatt = values_32[x] +vref*something
21:12.54makkonenmine's 0x4d0
21:13.01makkonencould be some sort of calibration?
21:13.10phhfor me, there is ~ 0.3V between values_16[x] and vbatt
21:13.12phhmakkonen: sound like so
21:13.28phhhehe
21:13.37phhmakkonen: could be battery capacity
21:13.47*** join/#htc-linux Neo31 (~Neo31@unaffiliated/neo31)
21:14.14phhhum not if it's called vref.
21:14.34makkonenreference voltage?
21:14.51phhhum I think so
21:14.54phhcould be wrong.
21:15.04makkonendon't know how that figures in, though.
21:15.17phhvref and vref 1/2 has some meaning for vregs
21:16.11makkonenand I'm still bugged by fc150 supposedly being discharge, but it just constantly reads 0x0000000d
21:16.54phhyou're reading it while connected to usb ? :p
21:17.26makkonenyeah, but it's the same way when I check dmesg from linux
21:17.36phhok
21:17.41phhI think it's most of the time 0 on rhod
21:18.56IceBoneHey, guys, a friend of mine wants to install Sense on his Magic, but the HTC site doesn't accept his serial as valid so he can't download it. Has anyone here done this and can send me the files?
21:19.29phhsense on magic oO ?
21:19.54IceBoneYeah. It was an official update.
21:20.27makkonenand temperature reads somewhere around 0xc10 no matter if the phone feels like ice or like the surface of the sun. basically I don't trust any of the values in the smem.
21:20.46IceBoneROM Upgrade with HTC Sense for HTC Magic
21:20.46IceBoneAn updated ROM is now available for HTC Magic customers to download. As always, HTC is committed to providing customers with the best possible mobile experience.
21:20.46IceBoneDate : 2010-01-2
21:21.20phhmakkonen: I've seen correct temp values only on diam100
21:21.25makkonenthey won't even let you download the rom without a correct serial#? that's pretty draconian.
21:21.27phhrhod has wrong data too
21:21.52IceBonemakkonen: and even looks like the system is borked as he tried entering it and it won't let him.
21:22.02IceBoneAnyone else have a Magic here?
21:24.10phhsuch a shitty device ? god no :p
21:24.45IceBoneYou're hilarious.
21:25.13IceBoneAnyway, formatting the sd card and copying over the backed up 02.04. files fixed android for me, even though I have no idea what went wrong...
21:25.18IceBoneWTF is system.img?
21:25.34IceBoneMarket still won't work, though.
21:26.17IceBoneAlso, Android on the Magic is miles faster than on the diamond... I want an android phoooone!!! *whine*
21:26.47phhIceBone: which is the size of your ramzswap on diamond ?
21:26.56IceBoneWhere do I check that?
21:26.59phhsetting it to a reasonable value could help a lot.
21:27.09phhsdcard/user/eclair.user.conf
21:27.13phhdisksize or something like that
21:27.51IceBonecc_disksize=100
21:28.55phhhaha ok.
21:28.57IceBoneShould I set it to 1/4 ram like it says?
21:28.57phhset it to 20
21:29.01IceBone20? ok
21:29.38phhor 27 if you prefer.
21:30.41IceBoneDo I have to change it in the eclair.user.conf~ file as well?
21:31.07phhno
21:31.26IceBoneOk, booting up.
21:31.50IceBoneI saw Babijoee mention something about fixing the Market... So my market not working isn't just my problem?
21:32.05phhyup.
21:32.13IceBoneWhat caused it?
21:32.20phhI think it needs google chat or something like that
21:32.35IceBoneI haven't updated to his new build yet.
21:32.44IceBoneIt started happening when you suggested I try donut.
21:32.52IceBoneThen I went back to 02.04. eclair.
21:33.01IceBoneSince then, nothing will download.
21:33.10IceBoneThat was almost a week ago.
21:33.19phhhum
21:33.28phhbabijoee's broken build is from yesterday
21:33.30phhor today
21:33.49IceBoneYeah.
21:33.57IceBoneMy market's been buggy since then.
21:34.12IceBoneI think I have to wipe it.
21:34.16IceBoneBut I really don't want to do that.
21:34.24phhI know that :/
21:34.37phhI already tried to clean only market stuff, without much luck
21:35.01IceBoneHoly crap, this IS faster! :o
21:35.21phhindeed.
21:35.31IceBoneAnd I touched the capacitive buttons during boot..
21:35.33IceBone*sigh*
21:35.38phhhaha
21:36.04phhsuch thing happen
21:36.07IceBonehits reset and puts the phone far away from hands reach.
21:36.16phhhow will you boot android eh ? :p
21:36.36IceBoneYozgatg's dual boot.
21:36.37ToAsTcfhphh: i have no clueas to why this isnt working for me. ive looked and compared all the files in /drivers/cpufreq and they are all the same as mine. thats our against ur source
21:36.38phhpfff
21:36.40IceBoneWell, auto haret launcher.
21:37.24*** join/#htc-linux balsat (~balsat@87.72.13.34)
21:43.33stinebdholy crap
21:43.39phhstinebd: ?
21:43.59stinebdphh: that mirror has burned through half of its traffic in the past day
21:44.06phhlol
21:44.11*** join/#htc-linux m3dlg (~m3dlg@bb-87-81-252-83.ukonline.co.uk)
21:44.41phhstinebd: it's time that I explain to you that I expected you to do the opposite :p
21:44.47phhmirroring your rootfs stuff to xland
21:45.10stinebd?
21:45.31stinebdthat stuff doesn't need a mirror :D
21:45.58stinebdalso the latest release doesn't seem to be on xland
21:46.28phhstinebd: if you explose your quota ... :p
21:46.33phhok the end of the month is soon
21:47.10stinebdi didn't know this many people were using android
21:47.18IceBonephh: will this make the phone respond a bit faster too?
21:47.31phhIceBone: no, I have to write something for that
21:47.48IceBonestinebd: nexus one alone sold almost 100k units in the first month. There's a lot of people androiding.
21:47.58phhIceBone: only ?
21:48.01IceBonephh: would welcome that bit of code with open amrs.
21:48.03phhtss such a great device
21:48.03IceBonearms
21:48.05stinebdwell it's not a nexus one release ;)
21:48.17IceBoneI know, it's well behind the iPhone or the Droid.
21:48.27stinebdanyway, i'm gonna push xdandroid 27.02 back to xland
21:48.34stinebdunless it was removed for some known reason
21:48.55phhstinebd: last stats I got, it was around 1TB of trafic in two weeks iirc
21:49.01phhand it was in december
21:49.09stinebduh oh :D
21:49.40phhseriously I don't want to sleep near the server of this mirror :p
21:50.04stinebdxland must be unhappy heh
21:50.20phhsee what's their job
21:50.47phhI don't think that such uploads changes things that much :p
21:50.58Unholydamn chile got struck by an earthquake =/
21:51.08stinebdyes. hawaii is waiting for the wave now
21:51.10phhUnholy: 10 hours late.
21:51.23Unholy9"45 to be exact
21:51.29phhUnholy: wow
21:51.31Unholy9:45
21:51.36Unholynot the earth quakle
21:51.37phhthe 10 number was totally guessed by my nose.
21:51.39Unholythe hoursd
21:51.41Unholylol
21:51.50Unholyeartquake was 8.8
21:51.52Unholystill high
21:52.00phhI don't want to be there :/
21:52.05Unholyme nether
21:52.09phhstill they got way less deads than the others one
21:52.09Unholyso far 64 death's
21:52.12phhhaiti ?
21:52.25phhI've already forgoten who it was /o\
21:52.31Unholyya haiti
21:52.48*** join/#htc-linux paran (~paran@h-87-78.A163.priv.bahnhof.se)
21:53.19stinebdmaybe i should open an s3 account for this server
21:53.41stinebdok xdandroid 27.02 reuploaded to xland
21:54.15phhstinebd: we have xland.cz for free ...
21:54.17phhbut as you want
21:54.26stinebdok
21:54.32stinebdfree :o
21:54.59phhstinebd: well it could be useful if we want autobuilder
21:55.07phhI mean full android autobuilder.
21:55.38stinebdmine is a vps, not well suited for something that big
21:55.48stinebdtoo much disk i/o
21:56.27phhglemsom and me have "real" servers
21:56.28phhat home
21:56.33phhso no big bandwidth
21:56.35phhbut cpu power.
21:56.39phhand i/o and everything
21:57.33stinebdi should see how many downloads are from north america and see if i can serve them and redirect other continents to xland
21:58.01*** join/#htc-linux houmles (~houmles@vodnik.dkm.cz)
21:58.10*** join/#htc-linux m3dlg (~m3dlg@bb-87-81-252-83.ukonline.co.uk)
21:58.12stinebdthat would be mostly for my enjoyment ;)
21:58.17ToAsTcfhphh: i found this though in /kernel/cpuset.c http://en.pastebin.ca/1814977
21:59.30phhwhat about it ?
21:59.40*** join/#htc-linux JesusFreak316 (~JesusFrea@pool-173-65-77-190.tampfl.fios.verizon.net)
21:59.59ToAsTcfhits a bit different
22:00.02*** join/#htc-linux m3dlg (~m3dlg@bb-87-81-252-83.ukonline.co.uk)
22:00.16stinebdphh: babijoee had some topaz sensors.so last night. is there a repo for that somewhere?
22:00.23phhstinebd: totally useless
22:00.35stinebddoesn't work?
22:00.36phhstinebd: there is only one sensors lib for all devices
22:00.42cr2_hm. leo .acdb and nexus .acdb have different format ;)
22:00.46phhat the same place as usual
22:00.57stinebdandroid-hardware?
22:02.49phhyes
22:03.11stinebdMrPippy: i added you to rootfs repo btw
22:04.44phhMrPippy: so we're just waiting for your commit :D
22:04.52phh(ok I think I'll go to sleep before)
22:06.45MrPippyyup about to push
22:07.05*** join/#htc-linux houmles (~houmles@vodnik.dkm.cz)
22:07.11phhI'm ready to pull.
22:08.30Markinuscr2_: what did the  .acdb do? It's like the soundpara?
22:09.25MrPippyok pushed
22:11.23phhchange in libhardware_legacy ?
22:11.26phhI hate that :'(
22:12.37cr2_Markinus: yes
22:13.45stinebdphh: i'll set up rootfs mirroring to xland later, thanks. now i'm redirecting requests for full android releases to xland so i can save some bandwidth ;)
22:13.57cr2_Markinus: we are missing one more thing. the lcd panel detection and spi init
22:15.42MrPippyyeah, the full path to wlan.ko and wlan_loader are hardcoded in libhardware_legacy
22:18.17phhfor donut I had a patch to clean that
22:18.32phhbut I think I lost this patch
22:19.07*** join/#htc-linux Unholy (~IceChat7@24.227.27.77)
22:20.58Unholycan i use the overcloking setting with the normal glemsome autobuilds?, or od i have to compile my own kernel?
22:21.19phhglemsom work
22:21.40GlemSomworks
22:21.49GlemSom:P
22:22.04Unholyah any tool to check the clock?
22:22.25Markinuscr2_: is the display direct connected or over microp like Topaz?
22:22.56phhUnholy: cat /proc/cpuinfo ?
22:22.58phhsetcpu ?
22:22.58cr2_Markinus: the spi controller ?
22:23.11cr2_Markinus: it's on qsd8250
22:23.17Unholyi do that in terminal?
22:23.31cr2_Markinus: the spi operation is like on nexus.
22:23.35ToAsTcfhphh: setcpu is an app to set the cpu
22:23.43phhToAsTcfh: and it gives the cpu infos :p
22:23.50cr2_Markinus: only the data depends on the panel id, and is different from nexus
22:23.55ToAsTcfhi dont believe this patch is userland though
22:24.03stinebd2164 hits, 4.33% of hits, 93757044kB served, 63.32% of traffic, filename /XDANDROID.27.2.10.7z
22:24.19Unholybut setcpu resets the clock to 528
22:24.23Unholywont let me get past that
22:24.35Unholyi have 610000 in the command
22:24.55bzothe glemson kernel won't let you overlock
22:24.59phhbzo: ?!?
22:25.03ToAsTcfhi wish i was u :P
22:25.13bzothat one relied on max freq config being changed
22:25.17stinebdit needs a config option changed
22:25.18Unholyah thats y its not working
22:25.27phhbzo: ah it needs setcpu.
22:25.32bzophh: however, once you push my new patch, it will work without config change :P
22:25.41phhbzo: I already pushed it ?!?
22:25.55ToAsTcfhso setcpu will work with this patch?
22:26.01GlemSomI'm using the htc_msm_android_defconfig in the tree
22:26.05cr2_hm. a8700060 | 0000a037
22:26.17bzophh: ? i don't see it on gitorious
22:26.24phhbzo: oops
22:26.26phhcommited not pushed :p
22:26.42phhstill waiting for my speaker change
22:26.58stinebdMrPippy: nice helmet ;)
22:27.12bzook, well once that is pushed and the glemson builds it, oc will work for everyone
22:27.16phhstinebd: lol.
22:28.12Unholybzo: nice one
22:28.36MrPippyhehe you recognize whose it is?
22:28.44phhMrPippy: I can't figure it out :(
22:28.46phhI know it
22:28.50phhbut no clue where it comes from
22:29.08ToAsTcfhbzo : including me?! :P
22:29.11phhhum
22:29.15phhnot when zooming on it
22:29.22phhToAsTcfh: na as usual wince >>>>> android
22:29.23phhNA.
22:29.57ToAsTcfh:(
22:30.03stinebdheh
22:30.14stinebdMrPippy: no i don't actually
22:31.41cr2_Markinus: my cpu is case 3. not in uboot source
22:32.16Markinuscr2_: Ohh, there are differences? I didn't know this . . .
22:32.49cr2_Markinus: run 'pd 0xa8700060 4'
22:32.55MrPippyhttp://www.abc.net.au/reslib/200709/r187890_702080.jpg
22:34.09Markinuscr2_: a8700060 | 0000a037
22:34.15*** join/#htc-linux matthewb57 (~matthew.b@CPE001c25b6ebde-CM001bd71cec5c.cpe.net.cable.rogers.com)
22:34.28Markinuscr2_: same ...
22:34.48cr2_ok
22:41.30cr2_Markinus: is 1GHz cpu
22:41.34cr2_<PROTECTED>
22:41.36cr2_<PROTECTED>
22:41.36*** join/#htc-linux tigerzen (~tigerzen@81-66-199-118.rev.numericable.fr)
22:41.37cr2_<PROTECTED>
22:41.39cr2_<PROTECTED>
22:42.12phhand frequency seems hardware locked :/
22:42.37phhbzo: I commited your patch
22:42.41phhand I commited rhodium's speaker
22:43.31bzophh: ok, thx
22:44.22MrPippywow theres a lot of code needed for the rhod speaker
22:44.26Unholyso now we can overclock?
22:44.32phhMrPippy: I wanted to do it cleanly
22:44.35MrPippyis this the speaker, or earpiece, or both?
22:44.38phhspeaker
22:44.40phhonly speaker :p
22:44.51phhMrPippy: double mic will be even longer
22:44.58MrPippyearpiece is still through amss?
22:45.05phhyes
22:45.21phhI haven't tested my patch at all though :D
22:45.39phhI should try if it doesn't crash at least.
22:45.51bzoUnholy: once the glemsom build updates
22:46.06ToAsTcfhphh: andf what will happen if i apply the oc patch?
22:46.09Unholybzo: ok thanks
22:46.12phhToAsTcfh: it won't apply
22:46.34ToAsTcfhi could try to use the table though
22:47.12bzoToAsTcfh: little of the oc code changed, it's mostly to fix freq tables for multiple devices
22:47.36phhbzo: and fixed jiffies ?
22:48.35bzophh: yes, that is the same as the patch we never pushed out
22:48.59ToAsTcfhso have yall even noticed a change in the way the builds are running or is it like a small difference?
22:52.51Unholymenus seem little faster
22:53.09cr2_Markinus: very useful table http://htc-linux.org/wiki/index.php?title=Leo_SMEM
22:54.21Markinuscr2_: ohh, yes. . .wow
22:54.22*** join/#htc-linux zemanm (~zemanm@CPE-65-30-176-154.wi.res.rr.com)
22:54.47cr2_-smd41 CS_APPS_DSP CS_M)
22:55.00cr2_on nexus smd_alloc_channel() cid 41 does not exist
22:56.10cr2_these are missing on leo
22:56.14cr2_#
22:56.16cr2_[    8.840393] smd_alloc_channel() cid=50 size=08192 'SMD_DATA5_CNTL'
22:56.17cr2_#
22:56.18cr2_[    8.840454] smd_alloc_channel() cid=51 size=08192 'SMD_DATA6_CNTL'
22:56.20cr2_#
22:56.21cr2_[    8.840515] smd_alloc_channel() cid=52 size=08192 'SMD_DATA7_CNTL'
22:56.28phhcr2_: leo still uses ppp ?
22:56.41cr2_phh: i think so
22:56.41phhor qmnet ?
22:56.44phhok
22:57.19cr2_phh: [    8.860778] tpa2018d1_init ?
22:57.33phhcr2_: ah N1 too ?
22:57.38phhI noticed acer liquid used it
22:58.01stinebdmy server is very happy now
22:58.04cr2_phh: yes
22:58.12cr2_Markinus: is it the same on leo ?
22:58.15cr2_#
22:58.17cr2_[    9.492523] read CFG0 = aa5400c0, CFG1 = 6746a
22:58.17phhcr2_: the sources aren't pleasant though
22:58.19cr2_#
22:58.20cr2_[    9.492523] CFG0 = e85408c0, CFG1 = 4745e
22:58.21cr2_#
22:58.23cr2_[    9.492523] CFG0: cw/page=3 ud_sz=516 ecc_sz=10 spare_sz=0 num_addr_cycles=5
22:58.24cr2_#
22:58.26cr2_[    9.492614] NAND_READ_ID = 5500bcec
22:58.29cr2_phh: for acer ?
22:58.36phhyes
22:58.45phhand it relies on userland
22:58.47cr2_base on codeaurora ?
22:59.06phhmmmm
22:59.07phhlet me chec
22:59.09phhk
22:59.44phhhaha
22:59.47phhit seems not
23:00.05phhwell the tree I have is mahimahi's one not codeaurora's one
23:00.13cr2_where can i get a tarball ?
23:01.28Markinuscr2_: sry, what is same?
23:01.52cr2_Markinus: nand chip
23:02.07phhcr2_: leo has tpa2018 too ?
23:02.12cr2_http://htc-linux.org/wiki/index.php?title=Leo_NAND
23:02.17cr2_phh: no afaik
23:02.28phhok
23:02.32*** join/#htc-linux sowfeast (~jeffehry@87-194-37-176.bethere.co.uk)
23:04.42sowfeasthello everyone - i've been trying to get my htcleo running with a bitbaked kernel - anyone know how to uncompress my zImage?
23:06.07tigerzenhéhé
23:06.33sowfeasti got your pm
23:06.35sowfeast;-)
23:06.52sowfeastgood to see you made it to the evtouch error!
23:07.07sowfeastthe fiz i suggested definitely works
23:07.11sowfeastfix
23:08.05tigerzeni looks for it but its rejected patch and its very past version
23:08.17tigerzenevtouch
23:08.32cr2_finding out what's wrong with the decompression is a better idea
23:08.43sowfeastyou have to download the file i mentioned, unzip the 04 and 04 diffs and put them where i suggested
23:08.56sowfeastthen update your bb file as i suggested
23:09.16phhcr2_: stop being logical.
23:09.23sowfeastyes, help with decompressing my zimage would be immensely helpful
23:09.43tigerzeni try it tomorow to late now then look here hardeware now al realesed http://www.htc-linux.org/wiki/index.php?title=Leo
23:09.47phhsowfeast: it's easy
23:09.50phhfind the gzip header
23:09.50sowfeastor did you mean fixing haret?
23:09.54phhextract what's next it
23:09.56phhand run gunzip
23:09.57sowfeastphh:  please tell me
23:10.14sowfeastwhat do you mean "what's next to it"?
23:10.24phhbah there is stuff before the header and stuff after
23:10.29tigerzengzip -dc *.gz | cpio -id
23:10.32sowfeastjust gunzip it right>
23:10.39sowfeastaha, i'll have a go, sec
23:10.49phhtigerzen: cpio ?
23:10.50phhfor a kernel ?
23:10.54phhyou're weird
23:11.05tigerzenoh sorry its for initrd ^
23:11.06tigerzen^
23:11.10tigerzen^^ !!
23:11.12phhtigerzen: no
23:11.13phhfor initramfs
23:11.29phhcr2_: sometimes ago, you asked how to extract boot.img, is that still releavant ?
23:13.05cr2_phh: yes
23:13.43phhcr2_: that's what I just described :p in the boot.img there are two times the gzip header
23:13.54phhfirst one is the kernel, second one is the initramfs
23:14.06tigerzensowfeast: its bin image
23:14.13sowfeastgzip: zImage-2.6.29-htcleo.bin: not in gzip format
23:14.15sowfeastgot that
23:14.39*** join/#htc-linux krisje8 (krisje8@gateway/shell/blinkenshell.org/x-mlkuhzarfukwbzfm)
23:14.56cr2_phh: how do i slpit it ?
23:15.10phhgzip header is 8b1f 0808 or 8b1f 0008
23:15.42phhcr2_: I do dd if=boot.img bs=<position of the header> skip=1 |gunzip -c > blabla
23:15.48phhgunzip removes the garbage
23:15.55phhthere might be some missing infos though
23:16.32cr2_phh: starts with ANDROID! and some cmdline params
23:16.51phhcmdline params are in boot.img not zImage ?
23:16.51phhhum
23:16.58cr2_and then comes something like arm asm
23:17.00phhI guess there might be some other infos in there
23:17.08phhcr2_: like... zImage decompressor ? :p
23:17.34cr2_msmsdcc_sdioirq=1
23:18.31cr2_phh: maybe
23:18.32phhcr2_: I think zImage starts at 04000
23:18.39cr2_then comes -- System halted
23:18.55cr2_the usual zimage
23:19.36cr2_0x4000 is in the middel of the gzip text strings
23:19.40phhna
23:19.41phh04000
23:19.44phhnot 0x4000 :p
23:19.48phhoctal
23:19.56Markinuscr2: here: dev0cfg0, dev0cfg1, dev1cfg0, dev2cfg1
23:19.56MarkinusHaRET(3)# pd 0xa0a00020 4
23:19.56Markinusa0a00020 | a8000800
23:19.56MarkinusHaRET(4)# pd 0xa0a00024 4
23:19.56Markinusa0a00024 | 0007000f
23:19.57MarkinusHaRET(5)# pd 0xa0a00030 4
23:19.57Markinusa0a00030 | aad4001a
23:19.58MarkinusHaRET(6)# pd 0xa0a00034 4
23:19.58Markinusa0a00034 | 002101bd
23:20.07phhcr2_: 0x800 if you prefer
23:20.19phhthe writesize of the NAND I think ?
23:20.25cr2_Markinus: add to http://htc-linux.org/wiki/index.php?title=Leo_NAND
23:20.54cr2_phh: 0x800 starts the arm asm
23:20.59*** join/#htc-linux randomblame (~randombla@c-71-59-184-96.hsd1.wa.comcast.net)
23:21.12Markinusok
23:21.15phhcr2_: headerless zImage ?
23:21.47cr2_phh: innovation ?
23:21.49leviathanmickey|tv: http://pastebin.com/6j4GTPL7
23:21.52leviathanany idea?
23:22.32phhcr2_: qualcomm innovation
23:22.39cr2_hehe
23:22.49cr2_phh: isn't it android ?
23:23.09phhhttp://www.qualcomm.com/innovation/index.html
23:23.11phhcr2_: oh right.
23:23.15cr2_phh: i can disassemble this blob, but is it really needed ?
23:23.22phhcr2_: don't think so
23:24.23sowfeasti just found a file called Image in ~/OE/angstrom-dev/staging/htcleo-angstrom-linux-gnueabi/kernel/arch/arm/boot
23:24.28sowfeast:-)
23:24.31*** join/#htc-linux Reefermattness (~MattGNM@c-76-18-79-170.hsd1.nm.comcast.net)
23:24.41sowfeastis that what we're looking for - twice the size of the zImage
23:24.49phhstinebd: your server is in a really bad shape :p
23:25.00phh150kB/s, 3 forced close
23:25.05stinebdphh: i'm playing with it for a few minutes
23:25.13phhstinebd: right
23:25.49cr2_<PROTECTED>
23:26.12cr2_phh: haha
23:26.15cr2_<PROTECTED>
23:26.17cr2_<PROTECTED>
23:26.25phhoO
23:26.44cr2_then comes some header
23:26.47cr2_<PROTECTED>
23:26.49cr2_<PROTECTED>
23:26.50cr2_<PROTECTED>
23:26.51cr2_<PROTECTED>
23:26.53cr2_<PROTECTED>
23:27.12stinebdphh: mod-geoip is not its friend ;)
23:27.24stinebdsegfaults :(
23:27.53cr2_phh: a lot of mcr p15
23:28.00phhwhich means ?
23:28.41cr2_phh: which mcr NetRipper was talking about ?
23:30.26cr2_phh: starting from here is somethign compressed
23:30.29cr2_<PROTECTED>
23:30.31cr2_<PROTECTED>
23:30.33cr2_<PROTECTED>
23:31.49phhstinebd: hum it seems your rootfs is broken
23:31.54*** join/#htc-linux bagpuss_thecat (~bagpuss_t@2001:41c8:1:5253::2)
23:32.00phhstinebd: missing /proc and /sys folders
23:32.33stinebdandroid init should be creating those
23:32.47stinebdgit doesn't keep empty directories anyway
23:32.47phhstinebd: udev relies on it
23:33.04stinebdudev on an embedded device? :P
23:33.05phhso you get no mmcblk0pX
23:33.16phhno ?
23:33.29stinebdseems like a waste of cpu
23:33.44phharg we don't use udev ?!?
23:33.51phhthere is something I miss.
23:33.54phhwho creates the devs ?
23:34.36phhah MAKEDEVS ok
23:34.54phhcould be just partition detection that is broken by the absence of /sys
23:35.00phhbut wait
23:35.09phh<PROTECTED>
23:35.10phhbaaaaahhhhhh
23:35.15phhmy brain :(
23:35.45stinebdlol those directories are in the rootfs repo actually
23:35.49phhstinebd: missing /sdcard folder
23:35.51stinebdi stuck .gitignore files in there
23:36.00phhtrue
23:36.04cr2_phh: 22 February 2010] [01:07:56] <NetRipper>        c7, c14, 0 clears all caches when they're ARMv7 compatible, but doesnt work on this cpu
23:36.12phhstinebd: but it doesn't get in ext2
23:36.20stinebdi don't support ext2 :P
23:36.41phhwhat's the rootfs.img FS ?
23:36.45stinebdi support git ;)
23:37.10cr2_phh:      e24:       ee071f1e        mcr     15, 0, r1, cr7, cr14, {0}
23:37.22cr2_phh:      e48:       ee07af1e        mcr     15, 0, sl, cr7, cr14, {0}
23:37.22phhstop speaking to me about stuff I don't understand :p
23:37.36stinebdi will try mounting the latest image
23:37.55cr2_phh: NetRipper said it does not work on this cpu ^^^
23:38.12phhah
23:38.19cr2_phh: and the leo boot.img has it at least twice in the decompressor
23:38.19phhhow is that even possible ?
23:38.29cr2_s/leo/nexus/
23:38.38phhokk
23:38.59phhstinebd: groumpf it also lacks /system
23:39.03phhall empty folders
23:39.12phhgenext2fs "forgets" empty folders ?
23:39.25cr2_phh: and the DGT is not stuck. it works very well in wince.
23:39.43phhcr2_: it's the smem irq position that causes the problem ?
23:39.47Markinuscr2_: Ok, dumped this values whitch we had in the nexus dmesg and put in wiki. .
23:40.04cr2_Markinus: is it the same chip ?
23:40.21cr2_phh: need to look what wince does with DGT
23:40.21Markinuscr2_: A other Nand ID ..
23:40.35cr2_Markinus: a bigger one ?
23:40.48Markinuscr2_: 4Gbit Hynix chip
23:40.58cr2_Markinus: there was a nand chip list in wiki ?
23:41.06Markinus0x5510bcad
23:41.17Markinuscr2_: sourcecode
23:41.26cr2_ok
23:41.40phhstinebd: boots wwwwaayyyyyyy better after creating those folders by hand
23:42.12cr2_Markinus: http://wiki.xda-developers.com/index.php?pagename=MSM_NANDID ?
23:42.51cr2_hynix_h8BES0UQ0MCP  0x1000  0x40  0x800  0x840  0xAD  0xBC  2  0  0  0
23:43.29Markinusahh, didn't know this pagae .. yes, I see
23:43.55Markinuselse if (flash_id == 0x5510bcad)  /* 4Gbit Hynix chip */
23:43.55Markinusmtd->size = 512 << 20;  /* * num_chips */
23:44.04Markinussourcecode ...
23:44.40cr2_0x1000*0x40
23:45.36cr2_awk '{printf("0x%x\n",0x1000*0x40*0x800)}'
23:45.38cr2_0x20000000
23:45.46cr2_Markinus: 512 ?
23:47.22phhMrPippy: upload your rootfs somewhere ...
23:47.30phhstinebd's generator doesn't seem perfect
23:48.19Markinuscr2_: yes, 512 <<20? ->0x20000000 ?
23:48.47phh[  459.294860] mmc0: Data CRC error
23:48.47phhyipi
23:51.34phh[  617.200378] _dhdsdio_download_firmware: dongle nvram file download failed
23:51.35phhhum
23:51.36stinebdi don't see why we need any of these directories or files anyway
23:51.44stinebdlinux should be able to boot on its own
23:51.53phhstinebd: we need /system to mount something on it ?
23:52.31phhand /sys is needed to know if you use a partitioned SD or not
23:52.35phh<PROTECTED>
23:54.18stinebdfor the record my generator is a piece of sh*t because i didnt know genext2fs exists :)
23:54.45phhyou're using dd+mkfs ?
23:54.51stinebdyeah :(
23:54.56phhhaha
23:54.59stinebdi rewrote it and now it works
23:55.12MrPippystill want my rootfs?
23:55.30phhMrPippy: if it works then no
23:55.37phhMrPippy: it needs wifi on in wimo ?
23:55.40MrPippynope
23:55.51phharf
23:56.04phhI get weird mmc errors
23:56.28phhlike the one I got with the TI chip on my diamond when I first plaid with that -_-'
23:57.29MrPippywifi is on mmc0, i've never seen weird errors from that
23:57.37phhok i'm just luckythen
23:58.01MrPippyi get those command timeouts from sd card (mmc1) on every boot though
23:58.21phhI get crc data error
23:58.25phhwhich is way worst than timeout
23:58.41MrPippyon which mmc?
23:58.41phhI still get them :(
23:58.44phhmmc0
23:59.09phh[   98.892059] mmc0: new SDIO card at address 0001
23:59.10phh[   98.914123] DHD: dongle ram size is set to 393216(orig 393216)
23:59.10phh[   99.066711] _dhdsdio_download_firmware: dongle nvram file download failed
23:59.10phh[  101.273193] dhd_bus_rxctl: resumed on timeout
23:59.10phh[  101.273254] dhd_preinit_ioctls: can't get MAC address , error=-110
23:59.10phh[  101.277465] eth0: Broadcom Dongle Host Driver mac=00:00:00:00:00:00
23:59.32MrPippydo you have a /data/wifi-nvram.txt ?
23:59.49phhhaha
23:59.50phhno.

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