IRC log for #htc-linux on 20100202

02:15.36*** join/#htc-linux apt (ibot@rikers.org)
02:15.36*** topic/#htc-linux is Welcome to htc-linux | Project homepage and wiki http://htc-linux.org | Livelogs: http://irclog.netripper.com/?chan=htc-linux | Logs: http://apt.rikers.org/%23htc-linux | htc-linux is not android
02:23.12adamwMrPippy: so i'm getting a bit lost - it looks like /etc/wifi has a script for loading the tiwlan driver as a module
02:23.31adamwMrPippy: and bcm4329 is apparently built in to the kernel but I can find no trace of it in the dmesg...maybe because of the missing firmware?
02:25.19adamwmmf, let's boot with the firmware in place and see what happens for now...
02:26.23MrPippycheck your dmesg for 'dhd', thats what the driver likes to be called
02:26.31adamwoh well that's intuitive!
02:26.52adamwsingle line - [   12.852020] dhd_module_init: sdio_register_driver failed
02:26.56adamwmight be the missing firmware i guess
02:26.59adamwi'll try with it in place
02:27.38MrPippymaybe, but i don't think so
02:27.55adamwyou think it's the gpios?
02:28.20MrPippymight need to set gpios in the driver too, but i'm not sure
02:29.51MrPippydebug/tracing can be turned on (at compile-time), but i'm not sure how
02:30.31MrPippyooh, tracing can be turned on at runtime
02:30.49adamwparameter?
02:31.53adamwyeah, same error with firmware in place
02:33.00MrPippyhmm actually yeah it needs a recompile, and a kernel param i think
02:33.32adamwoh feh i was hoping not to have to set up whatever build env it needs :)
02:33.53MrPippyits actually not that bad
02:34.13adamwis there an idiot's guide somewhere? i can build a kernel normally, dunno how the cross-compile or whatever it is works
02:34.31adamwi have the tree checked out already for reference
02:35.28MrPippyok heres the toolchain http://www.codesourcery.com/sgpp/lite/arm/portal/package3382/public/arm-none-eabi/arm-2008q3-39-arm-none-eabi-i686-pc-linux-gnu.tar.bz2
02:35.42adamwdl'ing
02:36.05MrPippydo you have the latest tree?
02:36.23*** join/#htc-linux MikeK (~dp@ppp121-44-247-112.lns20.syd7.internode.on.net)
02:36.47*** part/#htc-linux MikeK (~dp@ppp121-44-247-112.lns20.syd7.internode.on.net)
02:36.56adamwi cloned last night, i'll update now
02:37.35adamwok
02:38.07adamwoh crap - that toolchain looks to be i686?
02:38.10adamwi'm on x86-64
02:38.14adamwis there an x86-64 download?
02:38.37MrPippyyou can't run x86 binaries?
02:38.49adamwwell, depends how much environment they need i guess
02:38.55adamwif not much it should be OK
02:39.09MrPippyyeah its self-contained
02:39.19adamwah k should be able to make it fly then
02:39.43stinebdcross compiling the kernel is pretty easy due to the build system
02:40.01stinebdmake ARCH=arm CROSS_COMPILE=arm-none-linux-gnueabi- ...
02:40.03stinebdthe end
02:41.21adamwk, i'll give it a shot
02:42.10MrPippycopy the defconfig from arch/arm/configs/htc_msm_android_defconfig to .config first
02:43.19adamwlooks like i need i686 glibc at least, heh
02:44.07adamwso what do I need to change in bcm4329 anyway, pip?
02:45.04MrPippyin the makefile, add -DDHD_DEBUG to DHDCFLAGS
02:45.12adamwk
02:51.31adamwhmm, i'm not getting this right...
02:52.44adamwit wants a cc1 in the cross-compile toolchain but there doesn't seem to beo ne
02:53.03adamwhttp://fpaste.org/NLmv/
02:58.44MrPippyoh yeah, add arm-2008q3/bin to your PATH, then do make -j3 zImage ARCH=arm CROSS_COMPILE=arm-none-eabi- to compile
02:59.50adamwah that looks like the trick.
03:05.25adamwthen i need a kernel param as well?
03:06.44adamwaha that's fun. if you turn on the DEBUG param the module build fails :P
03:06.47adamwlet's see what they broke
03:07.52MrPippyhah why did i think that would happen
03:08.46adamwsome problem with a var called SRCBASE looks like
03:08.55adamwit's just comments that aren't terribly useful to us, i'll just pull them out
03:09.20adamwok, that fixes it
03:09.38adamwthat's the kind of coding I can do: 'mmmf, doesn't look important, rip it out'
03:09.39adamw:P
03:10.14adamwwhat do I do about modules? do I need a 'make modules' as well or can I just swap in this zImage and it'll find whatever it needs?
03:10.41MrPippyjust the zImage should work
03:10.44adamwk
03:11.30MrPippythe only things built as modules from the tree are the nand drivers (since you don't want those loaded except for testing)
03:11.36adamwk
03:12.05adamwso does it actually need a kernel param as well? or just with the modified zimage should be ok?
03:12.32MrPippyi think it needs a kernel param
03:13.56adamwdo you happen to know what?
03:14.12adamwi might be able to figure it out with enough random poking through the source but it wouldn't be fast :P
03:14.13MrPippyits called dhd_msg_level, so i think you would put bcm4329.dhd_msg_level=<something> into cmdline
03:14.40adamwgah, module_param , that's what the function's called. d'oh.
03:15.19MrPippyand for the something...try 3
03:15.26adamwyeah
03:16.27adamwalright let's see how badly thsi goes wrong, heh
03:19.00adamwwell, it boots
03:20.05adamwoh dear, i'm getting hit by the 'touchscreen is nuts' bug
03:20.20MrPippyoh yeah, the new touchscreen driver is in now
03:20.41MrPippyit won't work unless you calibrate it
03:21.14adamwwhich you need new userspace for
03:21.20makkonensee, don't you wish you were hosting that rootfs on your site now?!
03:21.29adamwoh well, i think i'm still dumping the dmesg to a text file anyway
03:21.32adamwi'll just reboot and pull that, heh
03:21.53makkonen...or you could just grab it from the ppcgeeks thread. or the wave.
03:21.57adamwmakkonen: AHAH I HAVE OUTWITTED YOU
03:22.10makkonencurse you
03:23.28adamwwell, that appears to produce no more info
03:23.49adamwso either i screwed up in one of any number of possible ways, or the debug stuff ain't working. or we got the number wrong.
03:25.21adamwhmm. i don't think that's the right dmesg. too long.
03:26.25*** join/#htc-linux Unholy (~IceChat7@24.227.27.77)
03:27.32MrPippyhmm yeah i don't know if the module param would actually work, we should just hardcode the msg level
03:28.08adamwyeah i'm not sure you can pass a parameter to a module that's been built in to the kernel
03:28.34adamwi think the bcm4329.blahblah cmdline parameter just tries to pass the specified parameter to modprobe
03:28.40adamwwhich isn't a whole lot of good :)
03:29.18MrPippyin dhd_common.c, dhd_common_init(), go from DHD_ERROR_VAL to DHD_ERROR_VAL || DHD_TRACE_VAL
03:29.56MrPippyit should work for things compiled in, but figuring out the name is tricky
03:30.19adamwhmm you said it likes to be called dhd, maybe the module name is dhd...
03:30.22adamwanyway
03:30.52adamwso:
03:30.53adamwdhd_msg_level = DHD_ERROR_VAL || DHD_TRACE_VAL;
03:30.54adamwright?
03:31.04MrPippyyeah
03:31.14adamwkk
03:32.14*** join/#htc-linux BHSPitMonkey (~stephen@pool-71-252-194-100.dllstx.fios.verizon.net)
03:32.14*** join/#htc-linux BHSPitMonkey (~stephen@unaffiliated/bhspitmonkey)
03:33.19adamwowch, i gotta run out for dinner - i'll be back and poke this some more later
03:37.29*** join/#htc-linux mastermerlin (~merlin@pD957D9FC.dip.t-dialin.net)
03:40.00*** join/#htc-linux mcdull (~mcdull@n219076081115.netvigator.com)
03:59.00*** join/#htc-linux mcdull (~mcdull@n219076081115.netvigator.com)
04:05.29*** join/#htc-linux __mart (~mart@125-238-44-108.jetstream.xtra.co.nz)
04:51.46tmztdhd?
04:52.42adamwtmzt: yup...what about it?
04:53.53*** join/#htc-linux jeremychang_ (~jeremycha@117.19.103.183)
04:56.23tmztChicago: !!
04:58.43tmztchic: freeswitch? can that do a gv/jaiku type setup pretty easily?
04:59.17adamwMrPippy: hum
04:59.46adamwMrPippy: I think that last idea got debug mode enabled - but it doesn't produce an awful lot of extra output at a quick glance
05:00.44adamwMrPippy: http://fpaste.org/VWpQ/ is the dmesg dump
05:01.43tmztdco: mick: xf86-video-msm supports 7k now?
05:05.28tmztmickey|zzZZzz: any reason not to disable reboot on panic?
05:05.55*** join/#htc-linux MrPippy_ (~pip@adsl-75-11-165-94.dsl.sndg02.sbcglobal.net)
05:06.02adamwoh, he's back :P
05:07.27ToAsTcfhtmzt: are u guys getting close to camera?
05:07.54adamwMrPippy_: did you see my stuff around XX:59?
05:08.21*** join/#htc-linux goxboxlive (~jrs@mail2.hjellnesconsult.no)
05:08.30ToAsTcfhtmzt: driod eris source was released by htc this morning
05:08.49ToAsTcfhor htc desire
05:09.03luke-jrToAsTcfh: full source?
05:09.09ToAsTcfhyep
05:09.17ToAsTcfhi just compiled
05:09.22luke-jrnice
05:09.23ToAsTcfhfor no reason
05:09.27luke-jrwell
05:09.29ToAsTcfhlol
05:09.30luke-jrcompile doesn't mean it works
05:09.32ToAsTcfhno root
05:09.34luke-jror supports everything
05:09.59ToAsTcfhyeah but it should. my heroc source from htc did
05:10.33luke-jrO.o
05:10.40ToAsTcfhnow i wish i could root my wifes desire
05:10.52ToAsTcfhfor the sake of braking it
05:11.02ToAsTcfhshes not happy
05:11.13luke-jrdoesn't get why Nokia gets all the popularity merely for advertising as open source when it really isn't, yet HTC doesn't even advertise the point and actualyl does it?
05:11.17ToAsTcfhwith my tinkering obsession
05:12.00ToAsTcfhhtc does it 4 months after they sell u the device
05:12.07luke-jrToAsTcfh: oh?
05:12.17ToAsTcfhlike sone assholes
05:12.20adamwluke-jr: what's un-open at nokia?
05:12.24luke-jrso they are infringing?
05:12.40tmztthe modem daemon, the sgx driver, if you mean n900
05:12.45luke-jradamw: most of the apps and misc important bits of the hardware support
05:12.57ToAsTcfhyep but by the time u can do something they release the source
05:13.00adamwah, iswym.
05:13.05luke-jrtmzt: you forgot the all-essential battery charger
05:13.32luke-jrwonders why HTC bothers to infringe instead of just complying and using it as a selling point
05:13.35adamwtmzt: anyway, you randomly exclaimed 'dhd' - wondering what was going on?
05:13.47tmztwhat is dhd
05:13.52ToAsTcfhmaejrep reverse engineered the gsm hero source to work on the cdma version before htc released the source
05:13.58tmztI comment on backlog sometimes
05:14.12tmztbut I don't see the responses until I scroll down
05:14.32adamwtmzt: it's what bcm4329 likes to refer to itself as (because logging as bcm4329 would be, oh, too obvious or something(
05:14.33ToAsTcfhheh
05:14.51adamwi'm poking blindly at the bcm4329 driver to see if I can get anything out of it on my rhodium
05:15.01adamwmrpippy was helping but he's idle now i guess
05:15.18ToAsTcfheh im out
05:15.27luke-jris guessing bcm4329 must be significantly different from the bcm he used to hack on <.<
05:15.35adamwluke-jr: what was that?
05:15.51luke-jradamw: I forget it's number; it's a MIPS SoC for cable modems
05:15.56adamwoh heh
05:15.58luke-jrits*
05:16.22adamw4329 is the wireless/bluetooth/some other bits i forgot probably chip in the rhodium, doesn't use tiwlan like others
05:16.26luke-jrah
05:16.50adamwso we were just seeing if we could get anything vaguely useful out of it, even useful debug info. but tough sledding so far
05:19.44adamwwithout debugging turned on in the driver, you get just:
05:19.45adamw[   14.926025] dhd_module_init: sdio_register_driver failed
05:20.03adamwafter adding the debug mode mrpippy found, all I can find that's extra is:
05:20.14adamw#
05:20.14adamw[ 3.148376] dhd_module_init: Enter
05:20.14adamw#
05:20.14adamw[ 3.158020] dhd_bus_register: Enter
05:20.18adamw#
05:20.19adamw[ 3.167144] Dongle Host Driver, version 4.218.161.0
05:20.20adamw#
05:20.22adamw[ 3.167144] Compiled on Feb 1 2010 at 19:09:06
05:20.24adamwwhich ain't exactly setting the world on fire
05:23.09*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
05:27.03*** join/#htc-linux gehzumteufel (~gehzumteu@cpe-76-93-100-70.socal.res.rr.com)
05:27.31adamwoh hey gehz
05:27.33gehzumteufeladamw you there?
05:27.34gehzumteufelhaha
05:27.35gehzumteufelYeah dude
05:27.36adamwyup
05:27.44gehzumteufelI just did the keymap thing you talked about onthe board
05:27.50gehzumteufelI noticed some quirky things though
05:27.55adamwyup?
05:28.05gehzumteufelA few keys have the same event ID
05:28.20gehzumteufelBut there is a slight difference in the data, but not the 4 digit ID you mentioned
05:28.42gehzumteufelAnd I have the RHOD210 btw
05:29.04gehzumteufelSo layout is VERY close to the WWE keyboard pic you linked to
05:29.13adamwgehzumteufel: that's the same for a few keys on the tilt2 actually
05:29.19adamwit needs addressing at kernel level but I can't do that :/
05:29.23gehzumteufelYeah
05:29.25adamwwhich keys have the same codes?
05:29.26gehzumteufelfigured as much
05:29.29adamwoh
05:29.31gehzumteufelup and left arrows
05:29.37gehzumteufelvol up and down
05:29.44adamwup and left is same on tilt2
05:29.56adamwvol up and vol down are OK on tilt2, though. what kernel are you on?
05:30.05gehzumteufelfrom yesterday
05:30.07gehzumteufeliirc
05:30.27adamwdo you have the exact id? just the zImage file name is enough
05:31.00gehzumteufelJust pasted the IDs via pm for the keys
05:31.16gehzumteufelDid any of them show up as emoticons?
05:31.26gehzumteufelis using trillian as his IRC client lol lazyness
05:31.37adamwno, irc don't work that way
05:31.43gehzumteufelzImage-modules-20100131_224534
05:31.51adamwit can't transmit emoticons as images, whatever your client displays 'em as, it only transmits text
05:31.56gehzumteufelI knew it didn't, but fuck I haven't really fucked with IRC in ages
05:32.06gehzumteufelI used to be a total IRC whore
05:32.35adamwgehzumteufel: can you check again with 20100201_185127 ?
05:32.49adamwthat has some fixes markinen added for rhodium from his own tree
05:32.58adamwit may do something about the volume keys i guess, probably not the arrows though
05:33.09gehzumteufelYeah
05:33.29gehzumteufelDoes he have a RHOD100 or 210 or what?
05:33.43adamwi'm not actually sure what he has
05:33.45gehzumteufelthe 210 is the least modified of all the modified ones
05:33.50gehzumteufelOr so it seems anyways
05:33.55adamwi think he may even have a diamond2 rather than a rhodium
05:33.58gehzumteufelAh
05:34.00gehzumteufelTopaz
05:34.10adamwoh, one other thing - make a note for any of the buttons that show a different event node from the keyboard buttons
05:34.19adamwall the keyboard keys should show up as one event node (event3 probably)
05:34.31gehzumteufelkeyboard were all event2
05:34.34adamwok
05:34.41adamwany hardware buttons show up as something else?
05:34.53gehzumteufelI didn't check
05:35.24adamwa few of the non-keyboard keys will probably be on a different node
05:35.50adamwfor the tilt2, the 'call', 'menu' and 'back' buttons are on the same driver as the keyboard, but 'end call', 'power', 'vol up' and 'vol down' are on the other driver
05:35.59adamw(which'll show up as some other event number)
05:36.07gehzumteufelah
05:36.20gehzumteufeltransferring new kernel now
05:36.59tmztrhod?
05:37.04gehzumteufelyes
05:37.06gehzumteufel210
05:37.17adamwtmzt: i'm doing keyboard maps for rhodium
05:37.25tmzttilt2 is same as rhod100 mostly
05:37.27tmztcool
05:37.30adamwdone tilt2 already as that's what I hve
05:37.38adamwgehz is helping with his model
05:37.53gehzumteufelrhod210 is REALLY similar to rhod100
05:37.58gehzumteufelvery few changes
05:38.04luke-jr>_<
05:39.02adamwgehzumteufel: can you find a pic of the 210 keyboard? just so I know what the Fn mappings should be?
05:39.23gehzumteufelYeah super easy
05:39.41gehzumteufelhttp://pdadb.net/imageview.php?file=t-mobile_touch_pro2.jpg
05:40.46gehzumteufelOr here is a much clearer one
05:40.47gehzumteufelhttp://www.blogcdn.com/www.engadgetmobile.com/media/2009/08/t-mobile-touch-pro2-review-10.jpg
05:41.04adamwoh yeah we have a winner
05:41.50gehzumteufelNow you see why the 210 has the closest resemblance to the 100 (100 = unbranded and sold direct through HTC)
05:42.19adamwyeah
05:42.19gehzumteufelOnly a few small changes really
05:42.47adamwidly wonders if the page up / page down keysyms will work
05:42.58adamwalright, i'll see what I can do
05:44.46gehzumteufelwtf the damn screen is fucked
05:44.51gehzumteufelin the new kernel
05:45.07gehzumteufelThe swipes are switched
05:45.12adamwgehzumteufel: yeah i told you not to get the VERY newest :)
05:45.19gehzumteufeloh lol
05:45.21gehzumteufeloops
05:45.22adamwgehzumteufel: i gave you the id for the last good one before the touchscreen changes
05:45.44adamwthe new code isn't wrong but you need a new userspace with a calibration tool apparently, haven't bothered with that myself yet
05:46.06adamw20100201_185127 is the one you want
05:46.50gehzumteufelAh yeah I have seen that calibration tool
05:46.58gehzumteufelBut never used it
05:48.28gehzumteufelIn the newest, the end key works
05:48.30gehzumteufelagain
05:49.27adamwyeah as I said a few of markinus' fixes (that were in the older kernel in matt's .cab build) have now gone into the upstream builds
05:49.49gehzumteufelFinally
05:49.55adamwso the end call key is back and the power button works for short press (home) instead of always thinking you're doing long press (task switch)
05:51.53gehzumteufelIf only they can get power management working
05:52.22gehzumteufelI am actually looking at getting the HD2 soon after it comes out. I just got this damn phone in August lol
05:52.24adamwgehzumteufel: hum, your Enter scancode is unusual - does the enter key work at present?
05:52.46gehzumteufelIt was working, yes. (I am in winmob to copy the slightly older kernel)
05:52.52adamwhum
05:53.10adamwoh heh lol
05:53.11adamwnever mind
05:53.16adamwi typo'ed in my hex converter :)
05:53.21gehzumteufellol
05:53.22adamwjust put 'c' instead of '1c'
05:53.31gehzumteufelah that will do it
05:53.55adamwthe codes you get out of getevents are in hex but the config file is in decimal
05:53.57adamwyou know, just for fun
05:54.05tmztafe you get up and down events?
05:54.05adamwwouldn't do to make it too easy
05:54.12adamwyeah
05:54.14tmztare you sending KEY_MISC also?
05:54.17tmztdoes fn work?
05:54.22tmztand u :)
05:54.27adamwfn gets mapped as alt_left
05:54.36tmztwhy that?
05:54.38gehzumteufelfn doesn't work the way it should. At least it wasn't in the old kernel I had
05:54.47adamwhow do you expect it to work?
05:55.00tmztto get the alternative keys to work it's best to use rightalt
05:55.00adamwthe  way I set it up you press fn and then press whatever you want and get the blue character
05:55.04adamwyou don't hold it down
05:55.08tmztand map the symbols with AltGr
05:55.46tmztso your doing the translation in the driver?
05:55.58adamwthe .kcm file does it
05:56.20MrPippyhi
05:56.25adamwtmzt: i just did alt_left because it worked
05:56.32adamwtmzt: what does alt_right do better?
05:56.33MrPippyi saw your dmesg, weird that it didn't print anything else
05:56.50adamwMrPippy: yeah, seems a bit odd - but otoh the debug mode is definitely kicking in or else it wouldn't have printed even that
05:57.14tmztsupports symbols
05:57.25tmztby mapping them as altgr charecters
05:57.28tmztis this android?
05:57.32adamwtmzt: yeah
05:57.32gehzumteufelYes
05:57.54adamwtmzt: well the way i've done it just uses the kcm files as described in http://pdk.android.com/online-pdk/guide/keymaps_keyboard_input.html
05:57.59adamwtmzt: seems to work perfectly well
05:58.22*** join/#htc-linux droid0011 (~g1@p4FDCA6D1.dip.t-dialin.net)
05:58.25adamwyou put whatever symbol's supposed to be produced by fn+key in the fn column of the kcm file
05:58.27gehzumteufelI think what he is getting at, is it is standard practice to use altgr for symbols
05:58.41gehzumteufelIt is standard even on PCs
05:58.54gehzumteufelJust not in the US
05:59.30gehzumteufelIn say, the UK, the right alt is the altgr key. It allows them to use the GBP symbol, amongst others
05:59.43adamwI know what it is
05:59.47tmztadamw: android isn't the only userland that should be supported here
05:59.50adamwis frm the UK in fact :)
05:59.52tmztcan it support either?
06:00.00adamwi haven't tried with alt_right
06:00.10tmztkernel keymaps really only work when something (a mod) can be mapped as altgr
06:00.13adamwi expect it depends what exactly android does with alt_right
06:00.15tmztand alt should be meta
06:00.22tmztor everything else breaks
06:00.32adamwwell i can do it with alt_right and see if it works for sure
06:00.35adamwno skin off my nose
06:00.42tmzthmm, what keysym are you using then?
06:00.53tmztthanks that would be helpful
06:00.57adamwuh? keysym for what?
06:01.07tmztkeycode constant for Fn
06:01.09adamwin the kl file you put keycodes, not keysyms
06:01.18tmztyep, in kernel
06:01.26tmztin the table in the kernel
06:01.37adamwoh. i didn't write that :)
06:01.37tmztthe raw keycode is too high
06:01.42tmztscancode I mean
06:01.47tmztthen you can't fix it
06:02.00adamwmarkinus is dealing with that side of things
06:02.04tmztI need KEY_RIGHTALT in the kernel
06:02.16adamwahh, ok, that's where we're getting crossed wires
06:02.21tmztthen you can tell android whatever keycode that matches that
06:02.32adamwright, i'm only doing the android level here
06:02.38adamwmarkinus is your guy for the kernel stuff
06:02.47tmztthis is until kernel upstream has serious support for the type of key boards
06:02.54tmztmark: ping
06:02.57tmztnot here
06:02.59adamwhe's not on yet i guess
06:03.04tmztplease pass the message :)
06:03.14adamwi'll just throw the log at him when I see him :)
06:03.30gehzumteufelYou can just message him
06:03.37gehzumteufeluse /memo
06:03.41gehzumteufelor on the board
06:03.43gehzumteufeleither way
06:03.44adamwthe hex scancode for Fn on gehz' keyboard is 01d0
06:03.50adamwlemme see what it is on mine...
06:04.00adamwsame.
06:04.03adamwso yeah, pretty high.
06:04.35gehzumteufelToo high for kernel?
06:04.39*** join/#htc-linux mcdull (~mcdull@n219076081115.netvigator.com)
06:05.00adamwgehzumteufel: it's just not what tmzt wants it to be...
06:05.09adamwoh, wait, that key's actually shift on my keyboard, fn on gehz's.
06:05.18adamwour shift and alt are flipped.
06:06.07gehzumteufelThat they are
06:06.25gehzumteufelAnd you have an OK button, whereas I do not
06:06.37gehzumteufelYou're "ok" button is my ctrl
06:06.57gehzumteufelAnd your home key is the envelope key
06:07.33gehzumteufeltitl2 keyboard blows
06:07.36adamwi might map my ok button to ctrl actually. ctrl would be more useful than an extra Enter. :P
06:07.41gehzumteufelhaha yeah
06:07.47adamwhah, mock all you want, at least I have a hardware forward slash :)
06:07.49gehzumteufelI LOVE being able to ctrl+c and all that
06:08.01gehzumteufelmeh, forward slash is hardly used
06:08.05adamwhaha
06:08.10adamwandroid is *linux* remember
06:08.14gehzumteufelI know
06:08.30gehzumteufelI have a hardware forward slash too
06:08.35gehzumteufelIt just requires the alt key
06:08.42gehzumteufelit is on the N button
06:08.49adamwyeah, i meant unmodified
06:08.54gehzumteufelAh yeah
06:09.00gehzumteufelBut you have to hold a modifier for ALL numbers
06:09.04gehzumteufelHow fucked is that
06:09.18adamw*shrug* hasn't bothered me
06:09.31adamwanyhoo'
06:09.42adamwi mapped alt+l to | (pipe) on the tilt2 layout
06:09.45adamwyou want that for yours too?
06:09.51adamwoh wait L is something else
06:09.54adamwi'll pick an unused key
06:10.00gehzumteufelyeah
06:10.04gehzumteufelWas just going to say that
06:10.11adamwany preference?
06:10.19gehzumteufelJust do a
06:10.21gehzumteufellol
06:10.42gehzumteufelI use pipe at work all the time
06:10.42adamwseems a bit awkward to two-finger
06:10.44adamwyeah that's why I mapped it :)
06:10.53adamwgiant pain in the butt having to get it off the software keyboard all the time
06:10.57gehzumteufeleah actually do Y or b
06:10.58gehzumteufelYeah*
06:11.00adamwk
06:11.10adamwerm, that is, OK :)
06:12.05gehzumteufelAlso, I am unsure, but how the hell do the ctrl+3 and ctrl+8 work to put on vibrate or put it in lock mode respectively
06:12.07*** join/#htc-linux kiozen (~oeichler@rgnb-5d87dad9.pool.mediaWays.net)
06:13.34adamwdidn't quite get that
06:13.42adamwyou mean right now, if you press ctrl and then 3, it vibrates?
06:13.50adamwhold ctrl and hit 3, or just press one then the other?
06:13.58gehzumteufelIn winmob, yes, it turns it to vibrate
06:14.06adamwoh i see
06:14.16adamwwell it's just using some kinda shortcut setup somewhere i guess
06:14.25gehzumteufel(technically it doesn't right now, because I have it set to WWE as opposed to TMO.RHOD layout)
06:14.31gehzumteufelYeah
06:14.41adamwi suppose you could set it to output some arbitrary keysym and then hook that up at a higher level in android
06:14.47tmztshift should be shift
06:15.00tmztor caps/shift on mine (rhod500)
06:15.07adamwtmzt: ?
06:15.19gehzumteufelIt is caps/shift on the 210
06:15.32adamwit's marked caps/shift on every rhodium i think
06:15.39tmztit's too high for linux userspace kernel keymap or at least for X
06:15.53adamwoh, i see. well as i said, depends on the model
06:16.06adamwon gehz's keyboard, the really high code (464) is the Fn key, not shift
06:16.12adamwon mine, the high code is shift, like for you
06:18.45tmzt464 is the problem whatever it is
06:18.57tmztwe don't have or want mtype's for all these devices
06:19.10tmztI think we can probe microp version
06:19.27tmzt464 is Fn on mine
06:19.33adamw464 is always the bottom left hand corner key
06:19.38tmztah
06:19.41adamwjust the label on that key differs between rhod variants
06:19.42tmztweird system
06:20.04tmzthow do you switch kcm based on device?
06:20.22gehzumteufelIn command line via startup.txt
06:20.39adamwactually the mapping of scancodes to physical keys is 100% the same between gehz' and mine
06:20.39gehzumteufelAt least that is what I presume from the stuff on xda regarding this
06:20.48adamwjust the labels printed on those keys is different
06:20.58adamwgehzumteufel: yes that's how we do it
06:21.07gehzumteufel:)
06:21.27adamwthere's some code in 'init' which reads /proc/cmdline and copies different .kl and .kcm.bin files into place depending on what you specify for the 'physkeyboard' parameter
06:21.45adamwgehzumteufel: you could actually use my tilt2 layout files perfectly well and pretend your keyboard has the tilt2 layout, hehe
06:21.53gehzumteufelLOL fuck that
06:21.54adamwor i could use yours, once i'm done with it, and relabel my keys :P
06:22.11gehzumteufelYou could swap keyboards and have a mostly good layout
06:22.13gehzumteufel:p
06:22.16adamwhehe
06:23.33adamwi could actually use a single .kl file for all rhodiums and just have .kcm files to remap the variant keys, but it would read very oddly
06:24.15gehzumteufelWhat do the kl files look like?
06:24.40gehzumteufelisn't going to download the repo
06:24.55adamwgehzumteufel: http://www.happyassassin.net/extras/tilt2
06:25.01adamwgehzumteufel: all the source files for the tilt2 mapping are in there
06:25.07adamwif you're curious
06:25.11adamw.kcm.bin is the compiled form of .kcm
06:25.46adamwthe microp-keypad.kl has most of the keys; navi_pad has the few that are handled by the gpio
06:27.08gehzumteufelWell then can't you just set it so that it has an "if physkeybd = tilt2, then reference tilt2.kcm" or whatever like that?
06:27.18tmztthey aren't gpio
06:27.29tmzt(not gpiolib I think)
06:27.42tmztalso, are they different on titlt2
06:27.53tmztignore missing buttons on other devices
06:27.58tmztare the keycodes the same?
06:28.09gehzumteufelThey probably are
06:28.10adamwgehzumteufel: that's exactly what it does
06:28.16gehzumteufelAh
06:28.20gehzumteufelWell then that makes sense
06:28.46adamwtmzt: if you look in init.etc/keymaps in the xdandroid rootfs.img you see a bunch of .kl and .kcm.bin files
06:29.20adamwand depending on what physkeyboard= is set to, different ones are copied to the 'active' names in /etc/keymaps in the actual root environment that's set up by init
06:29.55adamwif you look in the directory you can actually see the init file, i uploaded that too since I had to change it
06:30.01adamwlook right down the bottom and you can see how it works
06:30.09tmztphyskeyboard=(htccodename)?
06:30.29adamwtmzt: not entirely, there's ones like 'oldqwerty' and 'azerty'
06:30.34tmztuh, that's really messy
06:30.38adamwyeah
06:30.41adamwi didn't invent it :P
06:30.49tmztwho did and why?
06:30.50adamwone of those things that's just been knocked together as it went along
06:30.53adamwi guess
06:30.55gehzumteufellol yeah
06:31.17gehzumteufelI don't know how possible this is, but is there a way to detect what variant a phone model is?
06:31.30adamwi'm not how finely
06:31.47adamwyou can detect if it's a Topaz or Rhodium or whatever very easily i believe
06:31.57adamwbut not sure how easy it is to tell a RHOD100 from a RHOD210 etc programmatically
06:31.58gehzumteufelRight
06:32.05gehzumteufelYeah that is what I was referring to
06:32.19gehzumteufelSince that would make keymaps easy
06:32.30adamwi'm sure it can be done, probably one of those things someone should abstract out as a shared function somewhere
06:32.43stinebdtmzt: patches for better ways are welcome.
06:32.45adamwyeah, it could certainly be arranged better
06:32.46gehzumteufelYeah because that would actually make a lot of things easier
06:32.58gehzumteufelmtypes could be eliminated
06:33.24tmztthis is userland
06:33.38tmztif you want to differ kernel keycodes on these values
06:34.00adamwthere could theoretically be variants that weren't different to the kernel, i think...for e.g. it's would be possible for the only difference between my tilt2 and gehz' wing 2 to be what was printed on the keycaps
06:34.02tmztfor some like 464 which should be the linux keycode except for KEY_MISC which should be raw
06:34.16adamwthen there wouldn't necessarily be a way to detect it in the operating system
06:34.24adamwbut i dunno if there are actually any real world cases like that
06:34.34tmztI would suggest that and using htc cdoename xxxxnnn for for the value
06:34.48tmztthere are other azerty variants of different devices
06:35.04gehzumteufelyeah
06:35.24tmztas for identity
06:35.31tmztwe have something using seme
06:35.34tmztsmem
06:35.50gehzumteufeldoes that identify the different variants though of the same phone?
06:35.52gehzumteufelerr model
06:35.57tmztwe also can check the microp firmware version
06:36.01gehzumteufelSo like you have the rhod500 and I have the 210
06:36.12tmztyep
06:36.24stinebdhow would one detect a raph110 vs raph100?
06:36.39gehzumteufelAh then there should be an implementation in the kernel that passes that into the init to load the proper keymaps
06:37.07gehzumteufelThat can also be used to get away from mtypes
06:37.10stinebdideally the kernel would present proper scancodes and userland wouldn't need all these different layouts
06:37.16gehzumteufelagreed
06:37.20gehzumteufelBut this works for now
06:37.36gehzumteufelIt is a hack job at the moment, but it is better than nothing
06:37.41stinebdand i'm the one that came up with the init script hack because right now, there simply isn't anything else to do
06:37.55tmztthese things never get fixed
06:37.58gehzumteufellol
06:38.15gehzumteufelI am sure they will when it is a higher priority and more devs get in on it
06:38.24gehzumteufelObviously the biggest issue is drivers
06:38.26tmztwe still have people asking about cmdline options that sstarted out as workarounds
06:38.27stinebdi'd fix it if i knew of a way to
06:38.31tmztlike the ppp ones
06:38.53gehzumteufelstinebd: I didn't mean anyone specific, was just saying in general
06:38.55adamwwould still vote for someone to make sound work before fixing this :P
06:39.01gehzumteufelhaha
06:39.02adamw(not that we get to vote, i know)
06:39.04gehzumteufelwants sound too
06:39.18adamwi was trying to understand what's going on with making sound work earlier, but not getting anywhere fast.
06:39.20gehzumteufelAlthough, if I get the blackstone before it gets fixed, I won't care much lol
06:39.36gehzumteufelIn which then I will donate my rhod to the dev efforts.
06:40.04*** join/#htc-linux mcdull_ (~blackston@n219076081115.netvigator.com)
06:40.10adamwnice of you
06:40.30adamwmatt's planning to try and get a donation drive to buy a kernel dev a rhodium
06:40.50gehzumteufelIt's a little scuffed. No one will want to pay for it.
06:41.03adamwmeh, I got over $100 for my diamond with a crack in the casing
06:41.03gehzumteufelAlthough, I could just call up insurance and get a new one saying I lost the old one
06:41.11gehzumteufelserious?!
06:41.17adamwbest not say so in a public channel :P
06:41.20adamwwell, only a small crack, but still
06:41.33gehzumteufelmeh they aren't searching this shit
06:41.34adamwhttp://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=120522029938
06:41.48adamwcdma as well
06:41.56gehzumteufellol
06:41.57stinebdi wonder how much i could get for my old wizard
06:42.08adamwwell, i suppose some people want decent telus cdma phones to use with their cheap data plans
06:42.09gehzumteufelI sold my wizard LONG ago
06:42.19adamwhas a stack of titans heh
06:42.23gehzumteufellol
06:42.30*** join/#htc-linux playdo (~88020199@gateway/web/freenode/x-hfbwslrdaigblbvq)
06:42.31gehzumteufelI have a tytn sitting here
06:42.40gehzumteufelaka hermes
06:42.42adamwhttp://www.flickr.com/photos/adamwill/4314633878/
06:42.54gehzumteufeljesus
06:42.56gehzumteufelwhy so many?
06:42.58gehzumteufellol
06:43.00adamwmoto krzr, moto v360, diamond, titantitantitan
06:43.10adamwmeh they just accumulate
06:43.19adamwone of the titans went for a swim
06:43.20gehzumteufelheh
06:43.24*** join/#htc-linux leobaillard (~leobailla@leobaillard.org)
06:43.49playdogood morning guys
06:43.53*** join/#htc-linux kvaster (~kvaster@live.bn.by)
06:43.55adamwi'd donate 'em but I think everyone who wants to hack on titans has a titan hehe
06:44.14gehzumteufelYeah they are cheap now
06:44.27gehzumteufelreally cheap
06:44.37gehzumteufelMy tytn I couldn't get shit for it
06:44.43gehzumteufelI figured I would just keep it around for a spare
06:44.51playdoadamw, did u break some car windows with ur phones? :-D
06:45.09playdoespecially with the second one from top
06:45.22gehzumteufelhaha
06:45.51playdoand why the heck do u got 3 hermes? Oo
06:46.26adamwthey're titans...my partner sells cellphones for a living
06:46.28gehzumteufelThose aren't the hermes. They are the cdma variant.
06:46.33adamwafter a while spare ones tend to show up
06:46.42adamwtwo of the ones in that stack don't really work
06:46.51adamw(including the one that went swimming heh)
06:47.00gehzumteufelI love that HTC has quit making a totally different device for cdma vs gsm
06:47.14adamwcdma will be gone soon anyways
06:47.15gehzumteufelJust change what needs to be changed, but leave it as the same device otherwise
06:47.18playdomy hermes is a real extreme sprots fan ^^
06:47.29gehzumteufellol if by soon, you mean 10 years from now, yes.
06:47.33playdoit does swimming, skydiving etc.
06:47.49adamwgehzumteufel: well gone in the same way analog networks are gone now
06:47.53adamwi.e. not gone but we won't care :)
06:47.55gehzumteufelThe last analogue CDMA was only turned off in the US 2 years ago
06:48.08adamwtelus still has an analog network in canada i believe
06:48.31*** join/#htc-linux makkonen (~makkonen@cpe-66-69-229-9.austin.res.rr.com)
06:48.37adamwoh, no, they shut it off in 2008
06:48.39gehzumteufelAnalogue has been fully abandoned in the US. No installations that are operational currently exist.
06:49.07gehzumteufelSprint in the US is going to die a slow and painful death
06:49.15gehzumteufelTheir choice to use WiMax was a mistake
06:49.24gehzumteufelOne that will cost them their business
06:51.12playdojust played around with my hermes yesterday
06:51.18playdoflashed a new rom for fun
06:51.54playdogehzumteufel: did u try a linux on the tytn?
06:52.00gehzumteufelheh I haven't flashed mine in over 6 months.
06:52.03gehzumteufelI haven't
06:52.15gehzumteufelIs there an active project for it?
06:53.44playdodont think so, just xda-wiki entry
06:54.09playdowhats ur current phone? the hermes?
06:54.25gehzumteufelrhod210
06:54.41gehzumteufelYou?
06:54.46adamwgehzumteufel: did you get the vol up / vol down scancodes with latest working kernel?
06:55.07playdogot touch hd. which carrier do u got? O2?
06:55.07gehzumteufelshit I forgot and I had to reboot to charge my phone. Damn thing is almost dead
06:55.09adamwotherwise i'm just gonna guess it's the same as the tilt2
06:55.21adamwok i'll just go with the tilt2 navipad file for now
06:55.27adamwi'm 99% sure they'll be the same codes anyway
06:55.50gehzumteufelplaydo: TMOUS. And funny enough, I am thinking of getting the HD2 when it comes out here soon. It has 1gb rom. Twice the EU version.
06:56.59playdowhere are u from? Oo and why dp u got a german nick^^
06:57.15gehzumteufelLOL
06:58.01gehzumteufelBorn and raised in California. The German nick....I took German in secondary school. I have been using it ever since.
06:58.12playdocute ^^
06:58.21playdoi want 1gb ram too :-(
06:58.31gehzumteufelThe ram is only 576
06:58.42gehzumteufelBut the ROM (aka storage) is 1gb
06:59.17playdothats what i said :-P
06:59.19playdo^^
06:59.27gehzumteufelWhere are you playdo?
06:59.36playdowhy dont we get 1gb :( dammit
06:59.42gehzumteufelah I get ya
06:59.44playdoi'm in germany ^^
06:59.44gehzumteufelYeah that would be nice
06:59.54gehzumteufelOh nice
07:00.16gehzumteufelcr2 is there also
07:00.20gehzumteufelfrankfurt am main
07:00.23adamwgehzumteufel: can you grab http://www.happyassassin.net/extras/tilt2/rootfs.img and give it a shot according to my instructions?
07:00.28adamwgehzumteufel: it ought to do the job
07:00.29playdohe is german? didnt know
07:00.47gehzumteufeladamw yeah will check it out and report back once it is charged a bit
07:00.53playdofrankfurt is not far from here, just 250km or so
07:01.01adamwjust set physkeyboard=rhod210 , not physkeyboard=tilt2
07:01.06gehzumteufelk
07:01.13adamwi think I got everything right but chances are I nerfed something up, so test it out and let me know :)
07:01.21gehzumteufelplaydo just an hour or so away. easy drive or train ride
07:01.24adamwi'll boot it here and pretend mine's a rhod210 and see what it's like too
07:01.42adamwhad to do a bit more stuff with keysyms for yours
07:03.24adamwi think i hooked it up so the control key works, but...
07:03.39playdowhat time is it at ur location?
07:03.50adamwwho?
07:04.51gehzumteufel2304
07:05.22playdoi meant teufel. where are u from adamw?
07:05.33adamwvancouver canada
07:05.35adamw11:05pm here too
07:05.37gehzumteufelOh
07:05.47gehzumteufelMy flatmate is from Vancouver
07:06.04adamwoh cool
07:06.07playdohere its 8am and i have to work -.-
07:06.11adamwi'm _from_ england but I live here now
07:06.24gehzumteufelhah
07:06.33gehzumteufelI got a friend that is is in the midlands
07:06.52gehzumteufelAnd omg you guys have to watch this
07:06.53gehzumteufelhttp://www.youtube.com/watch?v=SXmv8quf_xM&feature=player_embedded
07:06.57gehzumteufelso funny
07:07.02gehzumteufelSince we're all nerds and all
07:07.10adamwReeferMattness: got a rootfs with rhod210 layout support now too i think :) just testing now then you can build a new cab, hehe
07:08.33adamwgehzumteufel: oh shoot haha I forgot to hook up the symbols for the number keys
07:08.38adamwreroll!
07:08.43gehzumteufellol good game
07:11.54playdoyou shouldn't let 12 year olds messing with pcs ^^
07:12.55gehzumteufelhaha
07:13.00gehzumteufelFunny shit eh?
07:13.05*** join/#htc-linux marcydarcy (~5b298cb3@gateway/web/freenode/x-uuekjyvnelriommd)
07:13.14gehzumteufelHe thinks only 5 people are connected to google
07:13.46playdodidn't really get that , cause i got no sound
07:13.53playdobut i read it ^^
07:14.18playdoare 5 people a lot?
07:14.19playdoROFL
07:14.41gehzumteufellol
07:15.07gehzumteufelHe goes on about a lot of dumb shit basically
07:15.24gehzumteufelHe says that "tracert" is how you see other people's IP address
07:15.37gehzumteufelAnd he says "tracer t" not trace route"
07:16.04adamwit's funny when you meet up with people and realize how many ways there are to say command names
07:16.04gehzumteufelAnd the ping numbers, he says that is the connection speed
07:16.05adamwheh
07:16.32gehzumteufelYeah but computer terms like trace route are pretty...ehh...uniform?
07:16.33gehzumteufellol
07:16.34playdo:-D
07:16.48adamwoh, yeah, traceroute is pretty unambiguous. heh.
07:16.56gehzumteufelhaha exactly
07:17.07playdosince every 12-year-old is playing counterstrike, latency and ping are connection speed xDDD
07:17.13gehzumteufelhaha yeah
07:17.18adamwgehzumteufel: ok, same URL, that one should work i think
07:17.25gehzumteufelHis youtube name is "nextgenhacker101" too
07:17.48playdogive him some time, maybe he got his master of it in 10 years ^^
07:17.52playdoIT
07:17.59adamwoh ffs I am full of epic fial tonight
07:18.04adamwmanaged to map both 9 and 0 to (
07:18.05adamw:P
07:18.17gehzumteufelhaha
07:18.18adamwyou can open your brackets but YOU CAN NEVER CLOSE THEM
07:18.19adamwmuahahaahha!
07:18.27gehzumteufelTime to get to bed for you adam
07:18.34adamwcower in fear of...the neverending expression!
07:18.51gehzumteufelhah
07:19.04playdoadamw is the devil haha
07:19.14playdothink u need some sleep ;)
07:19.34adamwpossibly :P
07:20.03playdodo u have some experience with netbooks/small notebooks?
07:20.09adamwgehzumteufel: ok aside from that it's working good. i think ctrl doesn't work, not sure why. and the fn+arrow keys may not work. but all the important stuff is good
07:20.13adamwplaydo: what kinda experience?
07:20.13playdowanna buy one in the near future
07:20.24adamwoh, as in, which to buy? yeah, some
07:20.30playdoyep
07:20.33adamwi have a vaio p
07:20.41adamwit is very cute and tiny and also slow as freaking hell
07:20.46playdoxD
07:20.46adamwatom is just...slow
07:21.02playdoi just need one for internet stuff and so on
07:21.09adamwif you have lots of money, get a vaio t or panasonic toughbook R series (you'll need to import from japan)
07:21.20adamwif you're *really* just gonna be browsing and emailing an atom is ok
07:21.21playdotry imagine me when i chat in the evening lying on my bed
07:21.25adamwbut you really had be sure that's all it's going to be
07:21.28playdoWITH AN 17" LAPTOP
07:21.58adamwotherwise the new breed of ~11", mid-price CULV laptops look nice
07:22.01gehzumteufelNetbooks are pointless
07:22.09gehzumteufelGet a 13"
07:22.27adamwacer has a range with 11.6" screens and CULV CPUs for around 400-700 CURRENCY (roughly)
07:22.30gehzumteufelUnless you just need a device that you can offload pictures and bang out a couple emails, they suck balls
07:22.32playdonono teufel, for my point, they are good. i need a bigger ppc with a keyboard ^^
07:22.32adamwthey look pretty nice
07:22.36playdoso netbook are perfect
07:22.43gehzumteufelGet an HTC Shift
07:22.52adamwas far as netbooks goes, there's really not a lot of play between 'em
07:22.55playdoi saw this one http://www.chip.de/artikel/Acer-Aspire-1810TZ-412G25N-Timeline-Notebook-Test_38366161.html
07:23.00adamwgo to a store where they sell a lot and find one who's screen and keyboard you like
07:23.05gehzumteufelI used to work for a computer manufacturer
07:23.21adamwin hardware terms, except for the oddballs like the sony p, they are all basically identical
07:23.27playdoi think shift is a small amount too small
07:23.38gehzumteufelplaydo what are you doing with it?
07:23.39adamwplaydo: that's exactly the 11.6" CULV range i was talking about :)
07:24.04playdoso there not much difference between the atom cpu and other components
07:24.20adamwplaydo: they're not officially netbooks, they're actually noticeably more powerful - they'll run like a low-end core 2 duo, which is pretty respectable
07:24.23playdogehzumteufel: just surfing, chatting and so on. for gaming i got mine 17"
07:24.46playdoadamw: yeah, and they are only slighly bigger, like 1" more or so
07:24.53adamwplaydo: i'd say that looks like a nice choice, i was considering the same model myself till i found out my budget would be rather nicer :)
07:24.58adamwyeah 13.3" is just a bit big for me
07:25.02adamw11.6 is a nice size
07:25.13gehzumteufelI have a 13". It is perfect
07:25.20*** join/#htc-linux StarLite (~nnscript@s55916cb1.adsl.wanadoo.nl)
07:25.27gehzumteufelThe machine I am on now is a 13
07:25.27playdojust "portable" enough for me, 13" is too much i thinl
07:25.48gehzumteufelheh I think you are underestimating the amount you will use it
07:25.57playdogotta rethink my budget ^^ but i dont want to spent more than 500€
07:26.01gehzumteufelTyping on the 11" notebooks suck
07:26.10playdowhy?
07:26.14playdosmall kb?
07:26.18gehzumteufelExactly
07:26.24gehzumteufelReally cramped and not comfortable
07:26.54playdomh
07:27.03playdogotta go to a shop and try it
07:27.14adamwgehzumteufel: ok, i've sent one final rootfs.img to my server, grab that, it should do all the important keys right
07:27.28playdotalking about feeling is useless, if i hasn't tried it yet
07:27.31adamwyeah, going to a shop is definitely the thing for laptops
07:27.39adamwscreen and keyboard are by far the most important thing about 'em
07:28.11playdoyeah, and i think i dont go for speed but more for usability/portability
07:28.23adamwgehzumteufel: let me know how it works out for you
07:28.26gehzumteufelAgreed, but like I said, I worked for a computer manufacturer. I have played with a good number of them.
07:28.56gehzumteufelTo each their own though. If you like it, great
07:29.01gehzumteufelDoesn't work for me though.
07:29.03adamwgehzumteufel: everyone's different, though. i can type fine on my vaio p, and that's tiny. it's not the size that bugs me about that system, just the speed
07:29.07playdogotta have to see the difference between 10" - 13"
07:29.44playdoi'm going to use linux on that one, so i think i can improve speed
07:29.55adamwplaydo: oh, mine runs linux. atom is slow no matter what OS you run
07:29.59adamwit's just intrisically slooooow
07:30.03gehzumteufelYes
07:30.04playdoand alter the system so it fits my needs and is very small
07:30.04adamwbut that aspire you're looking at would be nice and zippy
07:30.08gehzumteufelThe atom is a gimp system
07:30.17gehzumteufelerr cpu
07:30.35adamwno, you had it right the first time, the chipset and gpu options stink too :P
07:30.36gehzumteufelA 1.6ghz ATOM cpu, does as much as an 800mhz desktop cpu
07:30.37playdoi thought of that when i first saw the atom specs xD
07:30.45gehzumteufelhaha yeah I know
07:30.51gehzumteufelPinetrail helps a bit, but it still blows
07:31.08adamwat least it's not the bloody gma500
07:31.12adamwdon't get me started about that
07:31.16playdogoing to have a coffee
07:31.18gehzumteufel500 or 900?
07:31.20gehzumteufellol
07:31.21gehzumteufel:p
07:31.24adamw500
07:31.27gehzumteufelThey both blow major donkey balls
07:31.39adamwoh, 500 is in an entire different league of suck
07:31.42gehzumteufelThis machine has the 4500HD. Which is okay, but nothing to write home about
07:31.54makkonenyeah, but an atom at 1.6ghz runs like an 800mhz desktop chip... at 1.5 watts. It serves its purpose.
07:32.12adamwmakkonen: sure, you just have to be damn careful about the purpose :)
07:32.18gehzumteufelyeah
07:32.19gehzumteufelSeriously
07:32.24adamwi had to do my job for month on my P over christmas, that got wearing
07:33.09makkonenMy laptop's got a Quadro FX1600M.
07:33.18gehzumteufelnice
07:33.24makkonen...it's also pushing 8lb.
07:33.28gehzumteufelhaha
07:33.31adamw...and its own a/c system?> :D
07:33.33gehzumteufelCleo?
07:33.34playdo*gulp*
07:33.36gehzumteufelOr someone else?
07:33.43adamwalienware?
07:33.58makkonenand gets 2hr battery life if I'm careful.
07:33.59gehzumteufelPossibl
07:34.01playdothe worst thing are our pcs at work
07:34.01gehzumteufele
07:34.16gehzumteufelplaydo lol why?
07:34.36playdowe got 2.5ghz celeron, 256mb ram and full encrypted hdd
07:34.37makkonennah, not flashy. HP/Compaq 8710w. Great laptop.
07:34.37adamwoh jesus christ somebody please shoot me in the head
07:34.37gehzumteufelMy work pc is an athlon fx57 with 4gb of ram, 2x9600GT's and some other stuff
07:34.39playdoso image the speeeeeeed
07:34.42gehzumteufelplaydo eww
07:34.47adamwi think i've now got this wrong every way it's theoretically possible to get it wrong
07:34.56adamwhaving had them at ((, they're now at...)(
07:35.04gehzumteufellolol
07:35.05adamwooh! i didn't do )) yet! let's shoot for that this time
07:35.18gehzumteufeladamw time to give it a rest till tomorrow
07:35.27gehzumteufelYou're messing it up left and right :p
07:35.27adamwit's OK, i really genuinely can't possibly screw it up this time
07:35.30playdo:-D everythin wrong means u have just to say ~all
07:35.31adamwhehe
07:35.35gehzumteufelhaha
07:35.48adamwgehzumteufel: actually it's because the numbers are laid out 0123456789 in the kcm file
07:35.50makkonenplaydo: if only that were true.
07:35.56adamwso if you put ( first and ) second...they wind up the wrong way around
07:35.56playdo^^
07:35.57adamw*sigh*
07:36.16gehzumteufelah yeah
07:36.54adamwalright. this one's gonna be the charm.
07:37.16makkonenwe're all pulling for you.
07:37.21adamw=)
07:37.23makkonenWin one for the Gipper and all that.
07:37.27adamwhaha
07:37.29gehzumteufelhaha
07:37.34adamwdoes breathing exercises
07:37.43adamwvisualize...visualize...
07:38.29gehzumteufelwtf you use the US English spelling...I thought you were an Englishman?
07:38.44gehzumteufelwonders
07:38.54playdowoooohooo the hives are coming to an concert at the ultimative SAARLAND
07:39.08gehzumteufelSAARLANDBRUECKEN
07:39.09adamwgehzumteufel: meh, i talk to more americans than english these days
07:39.10gehzumteufellol
07:39.16gehzumteufelhaha I figured as much
07:39.22adamwalso my spellcheck is set to american, it yells at me if I type visualise :)
07:39.26gehzumteufelhaha
07:39.35playdosaarbruecken teufel ^^ do u know it?
07:39.54gehzumteufelIt is a state in Germany, if I remember correctly?
07:40.23gehzumteufelOr maybe a city?
07:40.26gehzumteufelIch fergesse
07:40.56adamwbtw, don't suppose anyone here knows what the *first* line of output from getevent means?
07:41.04adamwor can figure it out by looking at http://www.netmite.com/android/mydroid/system/core/toolbox/getevent.c ?
07:41.33gehzumteufeladamw in talking to all my UK and Canuck friends, I have unintentionally adopted the UK English spellings. And it has caused facebook to ask "would you prefer UK English?" I gave in. lol
07:41.35adamwmarkinus and I are slightly baffled by a few cases where two rhodium keys return exactly identical keycodes, but different first lines of getevent output. only we don't know what that line means.
07:41.42playdosaarland is the state, saarbruecken the capital. thats nice, u know more than some germans ^^
07:41.44adamwgehzumteufel: hehe
07:42.09adamwmost people in BC seem to use a mix of english and american
07:42.24gehzumteufelplaydo haha in my German class, we learned a lot about Germany in general. Not just the language. It was a full immersion. At least as immersed as you could in the US.
07:43.33gehzumteufeladamw I noticed on some of the ones that returned the same value, the 3rd value was the different one. So the 0067 as the same, but the last one showed 29 or 30 or whatever
07:43.46playdothats really deep and, not beeing rude, thats unlikely for us
07:43.48gehzumteufeladamw Yeah it has to do with the influx of tourists and shit
07:44.33gehzumteufelplaydo Yeah, it was fantastic. It was my most enjoyed and loved class during my secondary education.
07:46.45playdogehzumteufel thats cool, i don't think i learned that much abou the us. but we had a prof that was crazy about USA ^^
07:47.45gehzumteufelheh, there isn't much to learn about the US. It is a very insular, and unworldly place. No real identity
07:48.29gehzumteufelI grew up in a heavily Russian community. I, while born and raised in the US, have a hard time really identifying with most Americans. I just don't have the same views on things.
07:48.55gehzumteufelI have a more European view and thought process.
07:49.12playdothats great i think
07:49.27playdocause americans appear to be narrow minded
07:49.31gehzumteufelugh they are
07:49.53makkonenI don't know about the second part of that. Insular, not worldly, sure -- but there's plenty of identity. Somewhat variable throughout, but plenty... too much sometimes.
07:50.09playdoforbidding the evolution theory is just one example for americans acting stupid and keeping the people sort of dumb
07:50.29gehzumteufelWhat sort of identity does the US have? Aside from being a warmongering imbecile
07:50.55*** join/#htc-linux zachy (~george@ip-85-160-2-213.eurotel.cz)
07:50.58playdomakkonen thats true, at least afaik america seems to have a huge identity
07:50.59gehzumteufelplaydo: That is a sad fact, unfortunately. People here put more faith in god than science. And constantly want to force their views on others.
07:51.07adamwgehzumteufel: yeah that's what markinus and I saw
07:51.14adamwgehzumteufel: we just don't know what it means
07:51.18adamw(the 29/30 thing)
07:51.33gehzumteufeladamw: Yeah that is what I figured.
07:51.47playdowhat?
07:51.55gehzumteufelthe evolution thing
07:52.14playdoah k, wasn't with u any more ^^
07:52.43gehzumteufelhehe yeah
07:53.41gehzumteufelwhat is mark's sn?
07:55.12gehzumteufelplaydo: We do get DW in English here!
07:55.32adamwgehzumteufel: markinus , he's not online atm
07:55.53gehzumteufelapparently he isn't registered either
07:56.00gehzumteufel*** markinus There was no such nickname*** markinus End of WHOWAS
07:56.26adamwwhowas has never worked on freenode afaik
07:56.27gehzumteufelwtf
07:56.29gehzumteufelwhowas doesn't work
07:56.34adamwi've tried it on people who left 2 minutes ago
07:56.41gehzumteufelYeah I just tried it on cr2
07:56.46gehzumteufelshit don't work
07:57.11adamwyou know what's hilarious
07:57.18adamwi managed to screw up the keymap one more time
07:57.19gehzumteufel- Last seen  : Feb 02 02:43:32 2010 (5 hours, 13 minutes, 33 seconds ago)
07:57.24adamweven more hilariously - not in the way i predicted
07:57.27gehzumteufelthat is when mark was last on
07:57.31adamwi fixed it for shift+fn but not fn
07:57.31adamwhehe
07:57.34gehzumteufelhaha
07:58.32gehzumteufelNothing good to read on Reuters right now. Not that I want to read anyway.
07:58.46*** join/#htc-linux kiozen (~oeichler@p54920DC9.dip0.t-ipconnect.de)
07:59.40gehzumteufelWell, I am off. I need to sleep. See you all later
07:59.43gehzumteufeltake care
07:59.45playdosorry was afk gehzumteufel
07:59.55playdowhats dw?
08:00.03playdodevelopment ....
08:00.16gehzumteufelDeutsche Welle
08:00.30playdo:-D
08:00.30adamwchrist on a stick, finally
08:00.50makkonenyay!
08:00.52adamwgehzumteufel: one last time, http://www.happyassassin.net/extras/tilt2/rootfs.img
08:01.06playdodw music like kraftwerk and trio or what? ^^
08:01.06adamwgehzumteufel: that one's it
08:01.19playdo|afkafk another time
08:01.32gehzumteufeldeutsche welle the tv station
08:01.45adamwgehzumteufel: from my testing the important keys should all work as labelled. if the mappings for the call / menu / back / vol buttons are same as tilt2, they should work also
08:01.50adamwlet me know how it goes
08:02.02gehzumteufelk
08:02.29gehzumteufelWell I am really off. cheers wiedersehen ciao etc
08:02.32*** part/#htc-linux gehzumteufel (~gehzumteu@cpe-76-93-100-70.socal.res.rr.com)
08:05.35makkonenhmm. vilord's dumped the rootfs for vogue -- rolling it all into the system instead. seems intriguing.
08:11.17*** join/#htc-linux BabelO (~fcr@unaffiliated/babelo)
08:35.05*** join/#htc-linux balans (~Gebruiker@82-170-215-5.ip.telfort.nl)
08:54.37*** join/#htc-linux aatreya (~800cec2c@gateway/web/freenode/x-hvxgftxbtxfoylsf)
08:54.53*** part/#htc-linux aatreya (~800cec2c@gateway/web/freenode/x-hvxgftxbtxfoylsf)
09:13.36MrPippyi've brought in all the memory patches from the desirec source and trying 40MB from bank2, i'll leave it overnight and see what happens
09:32.53*** join/#htc-linux JoshuaL (~JoshuaL@cable-92-43.zeelandnet.nl)
09:37.42*** join/#htc-linux M1DLGpc (~M1DLGpc@bb-87-81-252-83.ukonline.co.uk)
09:47.31*** join/#htc-linux zycho (~zycho@f048153162.adsl.alicedsl.de)
09:51.02*** join/#htc-linux Markinus (~Miranda@gtng-4db043f4.pool.mediaWays.net)
09:52.09*** join/#htc-linux sxe (~quassel@ip-62-143-102-161.unitymediagroup.de)
09:52.55*** join/#htc-linux Taypen (prochat1@95.35.235.192)
10:07.32*** join/#htc-linux luminoso (~lumos@av-217-129-128-10.netvisao.pt)
10:07.35*** join/#htc-linux marcydarcy (~5b298bcf@gateway/web/freenode/session)
10:07.36*** join/#htc-linux marcydarcy (~5b298bcf@gateway/web/freenode/x-rbgwnqikhomztcul)
10:08.36*** join/#htc-linux StarLite (~nnscript@s55916cb1.adsl.wanadoo.nl)
10:09.18*** part/#htc-linux StarLite (~nnscript@s55916cb1.adsl.wanadoo.nl)
10:09.58*** join/#htc-linux StarLite (~nnscript@s55916cb1.adsl.wanadoo.nl)
10:12.09Taypenno one had problem with that?
10:12.33*** join/#htc-linux kvaster (~kvaster@93.84.112.80)
10:14.37*** join/#htc-linux GNUtoo|oeee (~GNUtoo@host46-130-dynamic.244-95-r.retail.telecomitalia.it)
10:17.22playdoproblem with what?
10:19.39Taypenwhen i run xandroid on my blackstone it tells "waiting for sd card"
10:20.29playdowhat sort of sd card do u have?
10:20.34playdoclass 4 and 6 got some problems
10:20.46Taypen4
10:21.04playdotry adding "msmsdcc_1bit msmsdcc_nopwrsave msmsdcc_fmax=14000000" to your cmdline (startup.txt
10:21.06playdo)
10:21.38playdoif that doesnt work get yourself a class 2 sdcard ;)
10:22.09Taypenbut second card is slower, no?
10:22.10playdoyour startup.txt should look like that:
10:22.28playdoyes, but if this workaround doesnt help, you need clas 2
10:22.35playdostartup.txt:
10:22.36playdoset ramsize 0x10000000 set ramaddr 0x10000000 set mtype 2030 set KERNEL zImage set initrd initrd.gz set cmdline "lcd.density=240 msmts_calib=0x395.0x3c9.0x79.0x77 clock-7x00.a11=500 msmvkeyb_toggle=off pmem.extra=1 gsensor_axis=-2,-1,3 msmsdcc_1bit msmsdcc_nopwrsave msmsdcc_fmax=14000000 mddi.width=480 mddi.height=800" boot
10:26.49*** join/#htc-linux GNUtoo (~GNUtoo@host46-130-dynamic.244-95-r.retail.telecomitalia.it)
10:27.09playdotaypen: is it now working?
10:28.40Taypensorry, im not at home right now but ill check that and tell u later, thanks for ur help
10:29.35playdook ^^ just try
10:29.49playdoor get an other sdcard on your way home
10:31.04Taypeni deleted the android thought that there ae no solution...
10:31.29playdojust read everthing, mainly www.xda-developers.com
10:31.51playdothere u get answers. just read the wiki, blackstone android thread and so on
10:33.19Taypenalready did, last hope was to check here :)
10:34.13*** join/#htc-linux Captnoord (~Captnoord@dc5147a47b.adsl.wanadoo.nl)
10:34.27Captnoordphh: i'm downloading your file now
10:34.35Captnoordbut I assume you can do the same
10:34.46Captnoordas rapidshare has lowered its defenses
10:35.55playdohey captnoord. how hast rs lowered def?
10:36.14Captnoordnah its a figure of speach
10:38.01playdocan u explain it? ^^
10:38.24playdodont get the connection between rs an that
10:38.51Captnoordwhen to many people are downloading from rs
10:38.54Captnoordit locks up
10:39.00Captnoordto supply only to paying people
10:39.26playdoand now it works without waiting?
10:39.29playdowhy is that?
10:40.50playdoah dont bother
10:40.56playdois going to lunch
10:42.13Taypenwhere can i get the kernel source, maybe the solution is just to select some different drivers?
10:43.01CaptnoordTaypen: read the wiki....
10:46.49Taypengood answer :)
11:18.59leobaillardfor those interested, I'm uploading a new build which contains the touchscreen calibration tool, the modified sensors lib for topa/rhod, the ledeffects program for diam/raph and the new tilt2 keymap
11:21.20*** join/#htc-linux StarLite (~nnscript@s55916cb1.adsl.wanadoo.nl)
11:44.34*** join/#htc-linux MethoS- (~clemens@134.102.106.250)
11:46.16playdo|lunchleobaillard, can i use the calib toll on blackstone?
11:48.16leobaillardplaydo: I'm not sure, according to the commit comment by phh, all board but blackstone are concerned, but that wasn't the last commit, so I'm not sure
11:48.36leobaillardbut nothing prevents you from giving it a try, it can't harm your device afaik
11:49.28playdoi know. i tried babijoee-wifi kernel from http://glemsom.anapnea.net/android/htc-msm-android/ for calib
11:49.37playdobut afaik it was just for diam
11:49.55playdoi give it a try when i'm at home and got the time
11:51.35leobaillardthe kernel only doesn't do the trick, you'll need my new build to test it
11:52.42playdosorry i meant the build on the given page, my bad
11:53.07playdo"To generate your own calibration file, you can use this build: babijoee-wifi.7z. (File will be placed in /sdcard/calibration.)"
12:12.29*** join/#htc-linux GNUtoo (~GNUtoo@host46-130-dynamic.244-95-r.retail.telecomitalia.it)
12:18.37*** join/#htc-linux IceBone (~IceBone@89-212-106-54.dynamic.dsl.t-2.net)
12:20.07*** part/#htc-linux IceBone (~IceBone@89-212-106-54.dynamic.dsl.t-2.net)
12:20.07*** join/#htc-linux IceBone (~IceBone@89-212-106-54.dynamic.dsl.t-2.net)
12:20.10IceBoneAnd now?
12:20.22IceBoneAaah.
12:25.48Markinusphh: HI, the bma driver is ready and it'S working. All is chacked in. Here is the patch for the sensors. We can now remove all of the old stuff. http://pastebin.com/m21f5fa6
12:30.16IceBoneUgh... weather widgets is completely useless... I don't know where it gets its info from, but 90% of the time it's inaccurate.
12:30.45leobaillardplaydo: the calibration you are speaking of is the wifi calibration file, not the touchscreen
12:31.39playdoand what does that calibrate? Oo
12:34.38leobaillardthe mac address
12:34.50playdo-.-"
12:35.11playdospent time for nothing :-D
13:00.47GNUtoohi, did someone look at msm-dri.c msm-dri2.c  msm-drm.h  msm-drm.c in https://www.codeaurora.org/gitweb/quic/xwin/?p=xf86-video-msm.git;a=tree;f=src;h=103c3ce48994fd96d8dfa802e840cacded95f022;hb=HEAD  ?
13:01.28GNUtoowhat does it do?
13:01.35GNUtoodoes it mean that we have 3d?
13:04.52*** join/#htc-linux c2d (~Mike@port-92-200-98-90.dynamic.qsc.de)
13:10.40*** join/#htc-linux philipgian (~philipgia@athedsl-4509418.home.otenet.gr)
13:15.09*** join/#htc-linux __mart (~mart@125-238-44-108.jetstream.xtra.co.nz)
13:18.01*** join/#htc-linux Squarc (~Squarc@82-217-32-29.cable.quicknet.nl)
13:19.12*** join/#htc-linux GeekLad (~GeekLad@adsl-178-231-40.jax.bellsouth.net)
13:21.11*** join/#htc-linux kvaster (~kvaster@93.84.112.80)
13:21.58*** join/#htc-linux Scepterr (~scepterr@ool-18b92277.dyn.optonline.net)
13:29.08*** join/#htc-linux dekar (~dekar@port-92-193-92-106.dynamic.qsc.de)
13:29.33*** join/#htc-linux marcydarcy (~5b298bcf@gateway/web/freenode/x-qjaotlrjsunvtarj)
13:54.28*** part/#htc-linux c2d (~Mike@port-92-200-98-90.dynamic.qsc.de)
13:59.16*** join/#htc-linux dcordes (~dcordes@unaffiliated/dcordes)
14:00.06dcordesGNUtoo ping
14:00.15GNUtoohi dcordes
14:00.32GNUtooI bet we need kernel with kms+etc... to get that working
14:00.47GNUtoobut as the howto says we could just disable 3d
14:00.59dcordeshowto?
14:01.02dcordesREADME
14:01.04dcordes?
14:01.33GNUtooyes README
14:01.36GNUtoosorry
14:01.46dcordesGNUtoo you need to use the 2.6.29 codeaurora branch
14:01.56GNUtoook
14:02.12dcordestry to run it on dream and se if you can get xorg to start
14:02.15GNUtootoo bad we use .32
14:02.21GNUtoook
14:02.35dcordesI spent 4 hours with this esterday night
14:02.41GNUtoook ouch
14:02.56GNUtoodcordes, so you have a recipe
14:03.00GNUtooI did a quick one
14:03.03dcordes:)
14:03.08GNUtoobut I didn't test it yet
14:03.12dcordesyep working one
14:03.18GNUtoostill compiling
14:03.19GNUtoook
14:03.32*** join/#htc-linux kholk (~kholk@unaffiliated/kholk)
14:03.40dcordesI will push it when I'm back home from lectures
14:03.55kholkhi all :)
14:04.17dcordesGNUtoo so it gets some more attention
14:04.30GNUtoook
14:05.10GNUtoois the 3d chip powerfull enough for playing extreme tux racer?
14:05.49dcordesI hope so
14:06.07GNUtooso it would be more powerfull than my desktop
14:06.53kholkgot a question.
14:07.05kholkI want to set an autobuild service for Leo linux kernels
14:07.10kholkbut I don't know how..
14:07.16kholkanyone can help? :)
14:07.34dcordeskholk it doesn't make sense at this point
14:07.44kholkheh, not at this point
14:07.53kholkbut in the nearly future it will make sense :P
14:08.21kholkI just want to set it and activate it when we'll need it ^^
14:08.54kholkand I'd be happy if someone can help me :P
14:10.16*** part/#htc-linux Dee2 (~dee@jane.earlsoft.co.uk)
14:13.11*** join/#htc-linux sektorNBA (ricardo@sektorNBA.org)
14:13.14sektorNBAhi
14:17.13sektorNBAwhat is the latest build of android to HTC diamond?
14:17.13dcordeskholk you can give me your email than we can get back at you soon as we need extra resources ok?
14:17.27sektorNBAi am at http://www.androidonhtc.com/diamond/latest_builds but i cant open the latest one
14:18.43playdoare u looking for stabel builds? or developing builds?
14:18.50kholkdcordes: giving it in query
14:19.45playdowhaks himself for bad spelling
14:23.52playdosektornba, are u away?
14:26.04playdoasking a question and then taking off *Grml*
14:26.42*** join/#htc-linux scheich (~philipp@p4FC5D02C.dip.t-dialin.net)
14:27.15scheichleviathan?
14:28.50sektorNBAno iam not. i was drinking water
14:29.11sektorNBAplaydo i want a -stable version.
14:29.39playdothen look into this thread http://forum.xda-developers.com/showthread.php?t=601751
14:30.01playdotake complete build from 24.01. and system.sqsh from 25.01
14:30.03sektorNBAoh xda. long time i dont join there
14:30.22playdoits very nice there ;)
14:30.30playdoso i gotta go home, work is ove
14:30.31playdocya
14:30.34sektorNBAbut i wanna Using one FAT32 Partition on SD Card
14:38.42*** join/#htc-linux sxe (~quassel@ip-62-143-102-161.unitymediagroup.de)
14:40.08*** join/#htc-linux kam187 (~kam187@81-179-8-102.dsl.pipex.com)
15:12.24sektorNBAcan anyone help me with HaRET? when i open it it ask a default.txt
15:12.31*** join/#htc-linux JoshuaL (~JoshuaL@kbl-tnz2380.zeelandnet.nl)
15:13.47IceBonedefault or startup
15:13.50IceBoneIt's in the STARTUPS dir.
15:14.06sektorNBAi saw
15:14.12sektorNBAi need to copy it to / ?
15:14.13IceBoneFind your version, copy it into root, where haret is, rename to startup.txt or default.txt
15:14.15*** join/#htc-linux graziano (~3e4d3815@fsf/member/graziano)
15:14.18*** part/#htc-linux graziano (~3e4d3815@fsf/member/graziano)
15:14.19sektorNBAk
15:21.43sektorNBAi am booting android at htc diamond and it is showing an cool X
15:22.06IceBoneI figured.
15:22.15IceBoneThe first boot up will take a bit.
15:22.20IceBoneSo it creates the data.img
15:22.35IceBoneAnd you better have disabled pin request in winmo.
15:24.44sektorNBAit loaded.
15:24.45sektorNBA;))
15:25.52IceBoneI updated to yesterday's kernel.
15:25.59IceBoneI should see if today's is any better.
15:26.26*** join/#htc-linux emwe (~michael@86.56.50.180)
15:27.06adamwGNUtoo: / dcordes: hey, X.org, something I actually know something about - fun. so msm is the native driver for our phones?
15:27.25GNUtoohi adamw
15:27.35adamwheya
15:28.00GNUtooadamw, https://www.codeaurora.org/gitweb/quic/xwin/?p=xf86-video-msm.git;a=summary
15:28.50adamwGNUtoo: so right now we're not using this at all, just using vesa or something? and the goal is - make this work so we can use it instead?
15:29.09GNUtooadamw, xf86-video-fb
15:29.18GNUtooadamw, yes
15:29.20adamwah, right.
15:29.28adamwhey, sounds like fun ;)
15:29.33adamwi'll maybe poke at it later
15:29.36adamwbeats working
15:29.39GNUtooadamw,I hope it will give us the following features:
15:29.45GNUtoo*xrandr -o 1
15:29.48GNUtoo*3d
15:30.00adamwhardware 2d acceleration too, right?
15:30.01GNUtoo*better performances
15:30.03GNUtooyes
15:30.30GNUtoo3d would need forward-porting some bits from another kernel version
15:30.44GNUtoobut if we get xrandr -o 1 that would be so great
15:31.19adamwum, -o 1 meaning screen rotation?
15:31.27GNUtooadamw, because for an unknown reason it doesn't seem to work on xf86-video-fb
15:31.29GNUtooadamw, yes
15:31.34adamwthat's from Ye Olde Xorg land, back when we still had xrandr 1.1
15:31.44adamwi would figure any driver being developed now would be an randr 1.2 driver
15:31.50GNUtoook
15:32.04GNUtoomaybe a miss-configuration then?
15:32.13adamwso it'd be something like xrandr --output OUTPUT_NAME --rotate right
15:32.15GNUtoos/miss/mis
15:32.33GNUtooadamw, xrandr -q doesn't show left,right etc...
15:32.42adamwon fb?
15:32.51GNUtooyes
15:33.04adamwyeah, that makes sense. fb is a Ye Olde driver :)
15:33.09adamwso it's still RandR 1.1.
15:33.17adamwbut if you get msm working it'll probably show all the left,right stuff.
15:33.46adamwi'm guessing, anyway. haven't looked at the code yet. maybe they are writing it as a 1.1 driver. would be kinda odd though.
15:33.51GNUtoohttp://pastebin.com/d2e2c198bd2e2c198b
15:33.56GNUtoook
15:34.02adamwempty
15:34.37GNUtoohttp://pastebin.com/d2e2c198b
15:35.17GNUtoook
15:35.31adamwthat's the fbdev config you're on atm i guess
15:35.36GNUtooyes
15:36.54adamwi'm guessing you're not testing with android here? just some kind of barebones boot which gives you a console to mess around at?
15:37.17GNUtooSHR
15:37.34GNUtooSHR is derived from angstrom
15:37.40GNUtoowhich is built with openembedded
15:37.52GNUtooSHR is focused on phones
15:38.01GNUtoothat's the distro to use if you want a phone call
15:39.51*** join/#htc-linux balans (~Gebruiker@82-170-215-5.ip.telfort.nl)
15:40.32adamwgotcha
15:43.29adamwwhat hardware are you on?
15:43.47GNUtoohtcdream
15:44.09adamwk. wonder if shr will run on my rhodium. one way to find out! :P
15:44.30adamware you cross-compiling the X driver? how painful is that?
15:44.56GNUtooI didn't cross-compile it yet....but dcordes did
15:45.07adamwah
15:45.53GNUtoobut if you have a working recipe and a lot of luck it's easy
15:45.57adamw=)
15:46.00GNUtoobitbake xf86-video-msm
15:46.14adamwoh yeah i see it in the scrollback...he's going to push it when he's back from lectures
15:47.57*** join/#htc-linux dcordes (~dccordes@unaffiliated/dcordes)
15:48.01adamwspeak of the devil
15:50.50GNUtoohi dcordes could you pastebin the recipe?
15:50.55*** join/#htc-linux kiozen (~oeichler@rgnb-5d87dad9.pool.mediaWays.net)
15:51.19GNUtooso I could have it bitbaken when I'll have the time to look at it
15:55.06*** join/#htc-linux Boydell (~Boydell@69.49.58.28)
15:57.22Boydellmorning all
15:59.17adamwhi
16:00.35*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
16:00.43Boydellany progress get made overnight?
16:00.57*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
16:01.06*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
16:01.17*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
16:01.28*** join/#htc-linux AstainHellbring (~AstainHel@unaffiliated/astainhellbring)
16:05.11adamw<leobaillard> 03:18:59> for those interested, I'm uploading a new build which contains the touchscreen calibration tool, the modified sensors lib for topa/rhod, the ledeffects program for diam/raph and the new tilt2 keymap
16:05.15adamw(that's xdandroid)
16:05.25adamwdon't see much about massive kernel breakthroughs :)
16:05.39Boydellbah "(
16:05.41Boydell:(
16:06.00adamw<Markinus> 04:25:48> phh: HI, the bma driver is ready and it'S working. All is chacked in. Here is the patch for the sensors. We can now remove all of the old stuff. http://pastebin.com/m21f5fa6
16:06.06dcordesGNUtoo, wait a sec
16:06.14GNUtoook
16:06.35adamwBoydell: and gnutoo and dcordes are playing with the x.org driver for msm
16:06.37adamwand that's about it
16:06.58Boydellah ok.
16:07.03GNUtooadamw, I am not playing with it yet...I need dcordes's recipe first
16:07.10leobaillardoh, about that, the build has been uploaded, I just haven't posted the link yet
16:07.11adamwmentally playing with it :)
16:07.12BoydellI dont see anything on the XDAndroid thread?
16:07.27leobaillardBoydell: see me previous message
16:07.32leobaillards/me/my/
16:07.37adamwleobaillard: I have a keymap for most non-tilt2 rhodiums probably incoming soon, getting it tested atm
16:07.41Boydellbetter than playing with something else ;) haha
16:07.59leobaillardgreat adamw
16:08.09adamwoh that's a neat bot trick.
16:08.58adamwsomeone was complaining that the power button didn't work, couldn't figure out why, then realized he didn't know it's *supposed* to be mapped to home :) heh
16:10.51emwetopa: dhd_module_init: sdio_register_driver failed?! (self-bulit linux-msm)
16:10.57emwesomebody else seen this? ^^
16:10.59adamwemwe: yes, me
16:11.05adamwemwe: that's the broadcom wifi driver not working
16:11.19adamwmrpippy was helping me poke it with sticks yesterday but we didn't really get very far
16:11.21emwehas no such hardware ;)
16:11.31*** join/#htc-linux luc_ (~luc@89-115-128-35.cl.ipv4ilink.net)
16:11.38adamwit's baked into the current kernel config so it probably just tries to load anyway
16:11.45emweyap, seems so
16:12.11emwewhonder if it's worth giving the recent kernel a try when the ts-calib-issue is yet to be sorted out
16:12.19adamwit is sorted out
16:12.24adamwyou need the new userspace with the calib util
16:12.28emweas i understand, we need to wait for the links being published, right?
16:12.30adamwleobaillard: where's that link? :)
16:12.33emwe;)
16:12.44emwei've seen a link in the glemsom comments
16:12.50emwenot looked at it yet, though
16:13.06adamwemwe: if you look in the drivers/net/wireless/bcm4329 tree in the source you'll see the dhd stuff in there
16:15.33adamwso the dhd/bcm4329 stuff on a phone *with* the chip currently works exactly as well as on a phone *without* the chip...excellent! :D
16:28.11sektorNBAi am getting Android keyboard ( com.android.inputmethod.latin) is not responding
16:29.36leobaillardhttp://htcandroid.xland.cz/XDANDROID.02.02.10.7z
16:29.56BoydellsektorNBA: I get that everytime I boot, just clock wait
16:30.08Boydell*click
16:30.28*** join/#htc-linux luc_ (~luc@89-115-128-35.cl.ipv4ilink.net)
16:33.10sektorNBAi clicked close
16:34.08Boydellok
16:34.42Boydellfor some reason, I think its with commit 31fb92c4d5a86d9b67cbdcf7eb2dd3d93ecf5d05
16:35.08Boydellmy boad-htcrapheal.c wont compile
16:35.32Boydellarch/arm/mach-msm/board-htcraphael.c:226: error: redefinition of 'touchscreen'
16:35.34Boydellarch/arm/mach-msm/board-htcraphael.c:152: error: previous definition of 'touchscreen' was here
16:35.36Boydellarch/arm/mach-msm/board-htcraphael.c: In function 'halibut_init':
16:35.38Boydellarch/arm/mach-msm/board-htcraphael.c:332: warning: passing argument 3 of 'msm_add_usb_devices' from incompatible pointer type
16:35.40Boydellarch/arm/mach-msm/board-htcraphael.c:334: warning: passing argument 3 of 'msm_add_usb_devices' from incompatible pointer type
16:36.43adamwBoydell: well look at lines 226 and 152 and see why it's getting defined twice
16:36.50Boydellyah I am now
16:39.19Boydellthat is down right stupid
16:39.28Boydellnot sure why it was defined twice
16:40.22adamwis one of 'em in some kind of clause that only applies to your hardware or smt?
16:40.39BoydellProb not. Its prob a bad merrge
16:41.41Boydellhappens again iboard-htcdiamond
16:41.51Boydell*in board-htcdiamond
16:42.35adamwwell there's builds on the site since that commit
16:42.38adamwso it obviously builds somewhere
16:42.50adamwthat's the commit that 'broke' touchscreen in all the builds people are whining about
16:43.01adamwso it may be something in your local tree i guess...
16:43.07Boydellit must be
16:44.27Boydellman there are ALOT of warnings when compiling
16:44.46adamwif it ain't full of warnings it ain't a good kernel build :P
16:44.57Boydelllmao
17:05.51Boydellleobaillard: Whats new in that package? the rootfs?
17:06.13leobaillardrootfs, squashfs and zImage (actually taken from glemsom's)
17:06.39leobaillardand the LedEffect app if you have a diam/raph
17:06.51BoydellBah. Ok. I was hoping the squash wasnt cuz that means I gotta copy shit over to my ext2 partition
17:06.51Boydelloh well
17:14.21*** join/#htc-linux onen|openBmap (~quassel@vbo91-1-89-87-201-85.dsl.club-internet.fr)
17:18.50adamwhuh.
17:19.00Boydell?
17:19.02adamwso i thought i'd try building bcm4329 as a module instead and see how it falls over when i load it
17:19.21adamwit complains about a couple of missing symbols - wifi_get_irq_number and wifi_set_power
17:19.35adamwexcept they are defined in board-htcrhodium.c so they shouldn't be missing, I guess?
17:21.11BoydellI dunno? I don't know much about building modules
17:21.22adamwobviously I don't either =)
17:21.28adamwwait for mrpippy I guess...
17:22.47adamwoh wait i may actually be able to do something about that
17:22.57Boydellleobaillard: something in that build breaks my phone. It wont boot. Gets to where is should load the boot animation, then resets back to WinMop
17:23.16leobaillarddid you try another time ?
17:23.24Boydelltwice
17:23.32adamwhmm, no, not what i thought.
17:23.59leobaillardBoydell: works for me. try it again. what's your phone model ?
17:24.18Boydellleobaillard: does this build also have the calibration app? its a raph800
17:25.03leobaillardyes, it has the calibration binary, I used it yesterday. let me double check though
17:25.15*** join/#htc-linux bzo (~chatzilla@netblock-68-183-234-14.dslextreme.com)
17:26.22leobaillardcheck ! the rootfs is correct
17:26.51BoydellIm re-coping everything over.
17:26.58BoydellI may be my zImage
17:27.00BoydellIm not sure
17:27.48*** join/#htc-linux f22 (~pcm_2@c-68-82-52-241.hsd1.nj.comcast.net)
17:29.19*** part/#htc-linux kam187 (~kam187@81-179-8-102.dsl.pipex.com)
17:29.54adamwooh. hmm.
17:32.55*** join/#htc-linux toi (~toi@d54C2A96D.access.telenet.be)
17:46.25*** join/#htc-linux houmles (~houmles@r6x23.net.upc.cz)
17:49.58BoydellOk, something in the lastest commits make my build bigger up.
17:50.03Boydellhow can I reset that?
17:50.09BoydellIve tried git reset --hard
17:50.15Boydelldoesnt go back far enough
17:50.18adamwno f'in clue, i suck at git.
17:50.37Boydellgit -revert
17:50.39Boydell?
17:50.55luke-jr...
17:51.19luke-jrBoydell: what exactly are you trying to do?
17:51.55BoydellUm, resest the head I guess. Something in the last merge to my local tree fucked up. I wanna go back I guess..
17:52.16luke-jrhow about finding what broke it?
17:52.32BoydellI have a feeling it was a bad merge
17:52.45luke-jrdefine bad merge
17:53.12Boydellit didnt merge properly?
17:53.44luke-jryou tried to merge it?
17:54.20Boydelli pull from a sub tree specific to my phone, the raph800, one in a while i pull from the main.
17:54.24*** join/#htc-linux BabelO (~fcr@unaffiliated/babelo)
17:54.32BoydellToday i pulled from the main, and thats when thing went wacko
17:54.59luke-jrso you want to reset to the raph800 branch?
17:55.26luke-jrgit reset --hard <branch-name>
17:55.37Boydellhmm
17:55.46Boydellis there a way to do that, but with the commit?
17:56.28luke-jr?
17:56.31adamwyay. i got a whole extra error message out of bcm4329
17:56.33adamwfeels proud
17:56.49Boydellso, reset to a commit number?
17:57.28Boydellin my log, i wanna reset to 13d90b528a897f3c7f4119af3c2d5ac9d36c07be
17:58.12luke-jrso put that
17:58.32*** join/#htc-linux luminoso (~lumos@av-217-129-128-10.netvisao.pt)
17:58.38Boydellthat worked. I wasnt sure if that would work or not lol
17:58.43*** join/#htc-linux GregLem (~greg@lal69-5-88-171-98-250.fbx.proxad.net)
18:01.43luke-jrtias?
18:03.13adamwi think bcm4329 has its own mmc handling stuff which is conflicting with whatever handles the mmc already...
18:03.33adamwhmm. bah.
18:03.36adamwneeds more pippy
18:08.25Boydellwe all need more pippy
18:09.23*** join/#htc-linux houmles (~houmles@vodnik.dkm.cz)
18:11.34*** join/#htc-linux balans (~Gebruiker@82-170-215-5.ip.telfort.nl)
18:14.09*** join/#htc-linux GlemSom (~glemsom@0x5da34bca.cpe.ge-1-1-0-1105.sdnqu1.customer.tele.dk)
18:15.32emwehm, something is "broken" after calibrating the screen ...
18:15.44Boydellby broken, you mean?
18:15.51emweleft right ok, top bottom seems inverted. (topa100)
18:16.13emwewhen i slide from top to bottom, at  ~ 20% from top, the app curtain is being pulled up
18:16.32Boydellweird
18:16.52emwesliding bottom to top, the notification area is not pulled "down" . seems the whole touch area is offset downwards and inverted.
18:17.02emwei've seen an error though ... let me dmesg...
18:17.53*** join/#htc-linux TimY (~timy@rue92-7-88-164-189-67.fbx.proxad.net)
18:19.19emwehm. dmesg only displays the "touch_calibration: point .." messages. but IIRC i saw an error when the points where about to echoed to the devs(?) entry...
18:23.29*** join/#htc-linux Captnoord (~Captnoord@dc5147a47b.adsl.wanadoo.nl)
18:26.03emwehm, ok. seems top-bottom is upside-down without offset.
18:40.05*** join/#htc-linux zycho (~zycho@g227168123.adsl.alicedsl.de)
18:43.30*** join/#htc-linux MrPippy (~pip@75.11.165.94)
18:43.53adamwMrPippy: ahah! just the man i wanted
18:44.29makkonenemwe: can you paste the dmesg lines with touch:calibration to me?
18:44.44makkonentouch_calibration, rather
18:44.57emweadamw: how about !machine_is_htcblackstone() in dhd_module_init() ?? ;)
18:45.43MrPippyhey there adamw
18:46.43adamwMrPippy: i think i may be getting somewhere v. slowly with bcm4329
18:46.50adamwmore details in a sec, my boss just wants something
18:47.01adamwbrb
18:47.28*** join/#htc-linux adamw (~adamw@redhat/adamw)
18:47.48adamwMrPippy: alright, so here's the deal
18:48.00adamwMrPippy: i tried building bcm4329 as a module for giggles, to see if that gave any more info
18:48.27adamwMrPippy: it complained that it was missing a couple of symbols - the ones added to board-htcrhodium.c
18:48.50adamwMrPippy: now I THINK the problem is that board-htcrhodium.c doesn't export those symbols
18:48.52MrPippyat compile or load?
18:48.55adamwboth
18:49.04adamwwarning when you do 'modules_install', error when you load
18:49.25adamwso anyhow I added EXPORT_SYMBOL lines and that gets rid of that error
18:49.33adamwnow 'modprobe bcm4329' takes quite a long time while it thinks about it
18:49.45adamwthen it fails with 'invalid parameter', and I get "Error: Driver 'bcmsdh_sdmmc' is already registered, aborting..." in teh log
18:49.49adamwwhich we weren't getting before
18:49.57adamwmy theory on that one is it's conflicting with the qualcomm mmc driver
18:50.02adamwi'm poking at that problem now...
18:50.18MrPippyhmm that could be true
18:50.47MrPippythe msm_sdcc is registering it as mmc0, but i'm pretty sure it has to
18:50.50adamwwhat i'm trying now is this:
18:51.09MrPippybcm4329 doesn't know how to work the sd controller by itself
18:51.20adamwin bcmsdh_sdmmc_linux.c there's a bit where it constructs the device support table
18:51.23adamwthat reads:
18:51.35adamw{ SDIO_DEVICE(SDIO_VENDOR_ID_BROADCOM, SDIO_DEVICE_ID_BROADCOM_4325) },
18:51.36adamw{ SDIO_DEVICE(SDIO_VENDOR_ID_BROADCOM, SDIO_DEVICE_ID_BROADCOM_4329) },
18:51.36adamw{ SDIO_DEVICE_CLASS(SDIO_CLASS_NONE)},
18:51.40adamwthat last line looks somewhat odd to me
18:51.54adamwi'm building with it taken out, just for the hell of it
18:52.54adamwyeah bcm4329 is obviously written to co-exist (it even has an ifdef where it does something different if the qualcomm driver is enabled)
18:53.00adamwbut yeah, anyway, that's where i'm at
18:53.23adamwi am now hoping you will wave your magic wand and fix everything ;)
18:54.18MrPippyhehe well i did find a sprint touch pro 2 on craigslist for $250, not sure if i'm gonna get it or not
18:54.54adamwMrPippy: you may wanna hang fire possibly - reefermattness is hatching a plan where we poor rhodium users pool donations to buy one of you guys a device
18:55.10adamwwe could possibly do it for less than that with a little, errr, creative insurance usage :)
18:55.20makkonen:-)
18:55.31makkonensprint does seem pretty willing to just give 'em out.
18:55.39adamwprobably got back stock
18:56.15MrPippyyeah they just zap the esn on whatever you replaced
18:56.17makkonenor possibly just hate supporting the original Touch Pro
18:56.31BoydellThey break all the time :(
18:57.53bzoI just wish the TP2 wasn't so darn big, probably the last decent phone Sprint SERO people can get
18:59.25MrPippyyeah i wonder if they'll let us have any eventual winmo7 phones
19:00.03makkonenTP2 seems like a perfect form factor to me. the fact that it's still got an effing 528mhz chip is a major tragedy, though.
19:00.34bzoI'm not optimistic about getting any new phones
19:00.52BoydellI cant wait to get rid of mine TP
19:01.14makkonenI'm sort of surprised they allowed the TP2 on sero. the whole thing doesn't make a hell of a lot of sense.
19:01.21bzoThe 528mhz is not so much the problem it's the poor IPC. I believe the new "budget" 600mhz chips like the palms have are much faster
19:01.41adamwmakkonen: yeah, it's quite silly. i only got it because I had to compromise on *one* of my requirements (ideal phone: telus 3G+euro 3G, hardware keyboard, snapdragon, android-capable, not a fricking lg)
19:01.58MrPippyits torture to get us to leave sero--"here, have this winmo phone"
19:02.00bzomakkonen: it does considering the absurd cost
19:02.02adamwacer liquid is the other closest candidate, but no keyboard.
19:02.51makkonenbzo: good point. except they keep giving them away for free!
19:03.00*** join/#htc-linux onen|openBmap (~quassel@vbo91-1-89-87-201-85.dsl.club-internet.fr)
19:03.05bzoMrPippy: lol, we'll have to thwart their plans with a good android port
19:03.52makkonenMrPippy: more specifically, "here, have this winmo phone that's exactly as slow and agonizing as the winmo phone you had 3 years ago."
19:04.14adamwbut it has a bigger screen!
19:04.24adamwoh, progress, how i love you.
19:04.44adamwfor extra bonus laughs, try playing video on the lovely lovely screen.
19:04.45bzowell as much as I am anti-iPhone, thank god it came out. It feels like winmo phones were stuck in the same spot for about 10 years until then
19:05.25Boydellbzo: I agree completely.
19:05.56makkonenadamw: wait what? can it not handle video?
19:06.24bzoMrPippy: how did you 2nd bank test go last night?
19:07.34adamwmakkonen: it's pretty creaky
19:07.44makkonenwhat a disappointment.
19:07.52MrPippyemwe: bzo: phone is still running with 147MB, 10 hours of uptime :-)
19:07.53adamwmakkonen: even with tcpmp under windows, it doesn't manage full frame rate at the full screen resolution in some videos
19:07.56adamwfor me anyway
19:08.12bzonice! :)
19:08.24emweMrPippy: grrr! you beat me ;)
19:08.42emweMrPippy: with which patches? pfn_valid *and* memmap_valid_within()? what else?
19:09.44adamwoh, the bcm4329 writer gets top marks for parameter name consistency
19:09.54adamwparmaster:           sd_msglevel:uint
19:09.54adamwparmaster:           dhd_msg_level:int
19:09.56adamw*doink*
19:10.40adamwoh haha it tab-completed parm:
19:10.50MrPippyyeah and another one-liner for include/linux/mmzone.h that i've found pretty much only in the desirec source
19:10.58emweMrPippy: whonder what i missed the last time i tried. but this actually makes me feel happy you got it. so i can drop phram experiments which actually hadn't worked out..
19:11.16emwewhat was it?
19:11.44MrPippyi'm only using 40MB from bank2 though, we'll see how far it can go
19:11.49adamwMrPippy: um. damn. just noticed something. i only get the 'is already registered' error the second or later time I try to load the module
19:11.57adamwMrPippy: the first attempt doesn't give that error
19:12.03*** join/#htc-linux playdo (~chatzilla@p57B3ADDD.dip.t-dialin.net)
19:12.08adamwso it looks like the first failure leaves it in unclean state and causes the error in later attempts
19:13.26adamwgrr. so we still just have a ten second pause and 'sdio_register_driver failed', sigh
19:13.43MrPippyemwe: -#define early_pfn_valid(pfn)   (1)
19:13.44MrPippy+#define early_pfn_valid(pfn)   pfn_valid(pfn)
19:13.50MrPippyfor include/linux/mmzone.h
19:13.58emweMrPippy: as nothing else seems to use bank2 (51MB have been at the beginning initially, now at the end) it seems unlikely, that the middle 26MB are used. so i'd go for using 128-51 and try ;)
19:14.28MrPippyok i'll be back in a little bit
19:15.15*** join/#htc-linux GregLem1 (~greg@lal69-5-88-171-98-250.fbx.proxad.net)
19:15.58emweMrPippy: damnit... had that too, but somehow screwed up... -#define early_pfn_valid(pfn)   (1)
19:15.58emwe+#define early_pfn_valid(pfn)   pfn_valid(pfn)
19:25.22*** join/#htc-linux dcordes (~dccordes@unaffiliated/dcordes)
19:28.10phh~ramconsole-diam
19:28.11apthmm... ramconsole-diam is pwf dm 0x00800000 0x00100000
19:30.57phh[01:22:43] <MrPippy> phh: your vbus patch yesterday broke usb for me, it would initialize ok but then go offline before init runs <----- I guess your smem vbus stuff is somewhere else, try to find where it is
19:31.03*** join/#htc-linux LeTama (~IceChat7@vau06-1-82-228-252-90.fbx.proxad.net)
19:31.24phh[01:42:22] <makkonen> MrPippy: this commit should fix it. http://gitorious.org/~makkonen/linux-on-qualcomm-s-msm/linux-msm-cdma/commit/cb0c6dd03ebbddaa2267a6a5810d302f612d2c16 <---- this should be done by userland
19:32.08*** join/#htc-linux Zoolooc (~fredsibar@p549559B9.dip.t-dialin.net)
19:33.30makkonenI disagree. we're changing the size of the calibration range in kernel, we might as well change the default calibration points on the screen as well. those values are general enough to work anywhere, and if you don't use them, then just feeding in calibration_points at boot time doesn't work.
19:34.16makkonen(feeding in calibration_screen and then calibration_points probably would work, but there's no reason to make extra work.)
19:34.30phhthe calibration_points alone has no meaning, you have to set the position of these points too
19:34.52phh(and yes the 2048 "screen size" is a hack I don't really appreciate.)
19:35.17phh(anyway, I can live with your changes)
19:35.37makkonenhow would you want to do the screen size?
19:35.43makkonenis there a non-hacky way?
19:36.01phhuse real hard resolution
19:36.21*** join/#htc-linux swc|666 (~infidel20@unaffiliated/swc666/x-4934821)
19:36.33phhiirc the values are 16bit
19:36.53*** join/#htc-linux houmles (~houmles@vodnik.dkm.cz)
19:37.06phhargh
19:37.12phhflashing from usb host I get "not allow"
19:39.13phh[03:36:48] <adamw> might be the missing firmware i guess <---- nop, firmware loading is after sdio registration
19:39.41adamwphh: yeah we don't think it's that
19:39.49adamwthat was my first ever guess :/
19:40.04phhhave you tried comparing N1's code and yours ?
19:40.06adamwphh: still don't know what the hell it _is_, though. i found the code for the error message we're seeing
19:40.14phhyou can also take a look at samsung i7500's kernel code
19:40.30adamwi was thinking of that but didn't know n1 code dump is available?
19:44.40adamwif i comment out the timeout in dhd_linux.c (the one that throws "sdio_register_driver failed" and gives up) then bcm4329 will load without error and show up in lsmod
19:44.48*** join/#htc-linux JoshuaL (~JoshuaL@cable-92-43.zeelandnet.nl)
19:44.49phh[13:35:47] <Markinus> phh: HI, the bma driver is ready and it'S working. All is chacked in. Here is the patch for the sensors. We can now remove all of the old stuff. http://pastebin.com/m21f5fa6 <------ pffff
19:44.55adamwbut no output in dmesg beyond the dhd init info, and no interface
19:44.59adamwso it obviously ain't working
19:45.12adamwsigh.
19:45.17phh[14:11:33] <GNUtoo> does it mean that we have 3d? <----- iirc it only means they use dri instead of pmem for memory management
19:45.19adamwphh: where's the nexus one code?
19:45.51makkonenphh: so the topaz has the touchscreen on a different device number (i.e. it's /sys/class/input/input1 instead of input3)... has tssc-calibrate.c been updated to take account of this? and, if not, what would be the best way to do so?
19:45.54phhgit://android.git.kernel.org/kernel/msm.git branch android-msm-2.6.29-nexusone
19:46.10*** join/#htc-linux c2d (~Mike@port-92-200-98-90.dynamic.qsc.de)
19:46.18phhmakkonen: yes it takes that in account, but the userland hasn't been though
19:46.24makkonenok
19:46.26phhwait a bit, I'll put it somewhere on git
19:46.31phhfirst i read my logs
19:47.04phh[15:17:05] <kholk> I want to set an autobuild service for Leo linux kernels <---- just ask GlemSom to make one :p
19:47.57phh[16:40:31] <GNUtoo> 3d would need forward-porting some bits from another kernel version <---- a free 3D driver for msm7k? I don't think so :p
19:48.09GNUtooouch :(
19:49.23phh[17:54:28] <Boydell> man there are ALOT of warnings when compiling <--- I hate it too :( make a patch ! :p
19:50.17phh[19:58:30] <adamw> MrPippy: it complained that it was missing a couple of symbols - the ones added to board-htcrhodium.c <------- if you want to do that you have to do EXPORT_SYMBOL
19:50.35adamwphh: yes, I did that bit already. see further dwon.
19:50.42phhright.
19:50.44phhdone reading logs
19:51.16adamwnotices that bcm4329 has two defines 'CUSTOMER_HW' and 'CUSTOMER_HW2' which are used all over the place to do things differently
19:51.21adamwpresumably according to different device implementations
19:51.23adamwlovely.
19:51.25*** join/#htc-linux lucx (~luc@89-115-128-35.cl.ipv4ilink.net)
19:52.49phhadamw: yes I love it too.
19:52.52phhI used the easier one to code :p
19:52.57*** join/#htc-linux lucx (~luc@89-115-128-35.cl.ipv4ilink.net)
19:53.07lucxphh are you
19:53.09lucx?
19:53.18phham I what ?
19:53.24adamwphh: what have you done with bcm4329? do you have it actually working in any case?
19:53.29makkonenare you. it's an existential question.
19:53.37adamwi am, and yet am not
19:53.53lucxwas you able to build wpa_suplicant from hostap for android?
19:53.56phhadamw: made it build, coded the board specific stuff
19:54.06phhlucx: yes
19:54.20lucxi have manage to build iw and libnl
19:54.20phhI don't remember how though
19:54.38adamwphh: k. i'm trying to make it do _something_ on this rhodium but not getting anywhere fast, heh.
19:54.44lucxbut i cant build wpa_suplicant
19:55.24phhadamw: yes i've seen the error you get, I opened the file, and I'll read and understand it when I'm done with dayly stuff
19:55.33lucxwpa_suplican from xandroid is from hostap or android sourcxes?
19:55.51adamwphh: thanks a lot, lemme know any thoughts you have - no rush, i'll keep plugging
19:56.09phhlucx: android
19:57.07lucxcan you help me to build wpa_suplicant from hostap
19:57.29phhno
19:58.07Captnoordwe'r here to share info... share work... if you don't have stuff done already....
19:58.08phhjust read what the web page says
19:59.06Captnoordbleh desquire doesn't handle 2 register variable stuff.... so umull would be a bitch to implement
19:59.07Captnoordlolz
20:01.35adamwis not going to take it any more and is going to write a script that does 'dmesg | grep -v BATT | grep -v "ena bit"'
20:01.35adamw:P
20:02.09adamwmeh, make it an alias.
20:02.17phhI've a better idea
20:02.20phhwait 2minutes.
20:02.30makkonenturn off the battery debug in the kernel?
20:02.37phhmake it optional
20:03.11makkonensmart
20:03.52phhoh it's already source optional.
20:03.58phhbah I prefer binary optional anyway
20:04.33adamwruntime option would be a lot more useful, yeah.
20:05.00adamwcompile-time logging options are just annoying.
20:05.51Captnoordnah they are usefull...
20:06.11Captnoordbut not in our situation
20:07.18CaptnoordMLAS    R3, R4, R7, LR  ; r3 = r4 * r7 + lr ( signed ) -> 80063040   R3 = R4 + R7 * LR;
20:07.19Captnoordlol
20:07.22Captnoordsomething goes wrong
20:11.13lucxphh when you builded wpa_suplicant you linked openssl ?
20:11.22phhdon't remember
20:11.47phh(seriously it was months ago)
20:11.55lucxok
20:12.49adamwphh: the nexusone diff looks interesting indeed; it's not a complete redesign or anything, but it has some changes right in dhd_linux.c, right around where we're having problems...seems like it moves the init of the wifi slightly earlier, adds some more error checking
20:12.58adamwphh: i'm just going to go ahead and try it with that version
20:13.11phhadamw: well I copied the driver code from nexus one :p
20:13.15phhbut I mean the board part
20:13.19adamwah ok :)
20:13.23adamwfrom an earlier version i guess?
20:13.29phhyes
20:13.53adamwah ok, i'll look at the board stuff too then
20:14.36adamwoh hehe
20:14.46adamwfrom the git log it looks like the change is designed to fix exactly what I ran into
20:15.01adamwthe resources not being released when it fails, so you get the extra error on subsequent attempts
20:15.12adamwso it'll stop that happening but probably still not work, hah.
20:16.41*** join/#htc-linux cr2_ (~cr2@ip-109-84-183-102.web.vodafone.de)
20:17.19playdohey cr2 :-) hab gehört du bist auch deutsch ;)
20:17.29playdohey to the rest ^^
20:19.17*** join/#htc-linux pipix (~buglabs@firebug.buglabs.net)
20:19.54adamwmeh. still not getting anywhere. i should do some work.
20:23.17*** join/#htc-linux TimY (~timy@rue92-7-88-164-189-67.fbx.proxad.net)
20:23.20*** join/#htc-linux cr2_ (~cr2@ip-109-84-183-102.web.vodafone.de)
20:24.18cr2_playdo: this channel is english though. since its very beginning. i quick look at the user ip will tell you where i'm from anyway
20:24.23cr2_phh: ping
20:24.27phhpong
20:24.37cr2_phh: new gps ideas ?
20:24.46phhnop
20:24.48cr2_the evil puzzle
20:24.51phh:p
20:25.21cr2_do you have gps redirected to a COM ?
20:25.38phhuh ? on wince you mean ?
20:25.46cr2_yes
20:25.52playdocr2 didn't see ur ip in the beginning, but u not surely can tell if the one behind is really from there
20:25.53phhCOM4 iirc yes why ?
20:26.06playdoat work, i appear to be in michigan^^
20:26.38cr2_playdo: i will use tor, if i was so cautious
20:26.49cr2_phh: yes, but you can use wince api
20:26.55cr2_phh: without any serial portt
20:27.09phhok (for remainder, I know pretty much nothing about wince.)
20:27.20MrPippyok here goes nothing, Memory: 107MB 77MB = 184MB total
20:27.26cr2_phh: ok
20:27.35phhMrPippy: ok so it's time for me to get a raph or rhod then :p
20:27.53cr2_phh: we can add a wince command, to open $1 and read 1024 bytes. or something like that
20:28.06phh$1?
20:28.30cr2_READ <devname> <nbytes>
20:28.49cr2_rather generic
20:29.17cr2_i wish we could teach haret to parse ioctl parameters
20:30.01cr2_then you can even read calibration data from nand
20:30.57MrPippywhat kind of calibration? g sensor, wifi?
20:31.02cr2_phh: roadmap tries to identify gps serial port
20:31.12phhMrPippy: both
20:31.12cr2_MrPippy: both
20:31.15cr2_lol
20:31.21phh(nananereuh.)
20:31.30phh(sorry I don't know how this is supposed to be translated :p)
20:31.44makkonenjinx?
20:32.37cr2_phh: but we need some smart way to pick the rpc message
20:32.50phhmmutrace isn't enough ?
20:32.50cr2_simple mmutrace will spam
20:32.54phhok.
20:33.02phhmmutrace + modified haretconsole ?
20:33.07cr2_you can watch the head/tail
20:33.13phharf
20:33.24phhhum this might be scripted though
20:33.25cr2_but they may be set by arm9 ?
20:33.42*** join/#htc-linux Soul_Shadow (~no@c-174-48-168-84.hsd1.fl.comcast.net)
20:33.45cr2_yes, you can buffer the rpc fifo, and try to parse it
20:33.49cr2_in python
20:33.52cr2_hehe
20:33.55*** join/#htc-linux Boydell (~Boydell@69.49.58.28)
20:34.01cr2_are you a great python person ?
20:34.20Soul_Shadowuestion anyone know a simple linux distro with a smooth GUI & easy installs? I need to run broadwave & ubuntun sucks bad
20:34.37phhcr2_: my scripts are done in C -_-'
20:34.38cr2_pH5 had even writen an arm disassembler in python
20:34.59emweMrPippy: nice that it works! congrats!
20:35.05cr2_yeah, i don't like highlevel languages very much.
20:35.10cr2_only awk :)
20:35.25cr2_which is interpreted c for me anyway.
20:36.11makkonenphh: can you upload/pastebin/whatever tssc-calibrate.c?
20:36.32phhI said I'd do a git !
20:36.32phhbaah
20:36.39makkonenoh. git is good.
20:36.50phhyes but it's) far in the todolist :p
20:36.54phhhttp://en.pastebin.ca/1776177
20:38.29*** part/#htc-linux Soul_Shadow (~no@c-174-48-168-84.hsd1.fl.comcast.net)
20:42.08playdowhat about calibration for blackstone?
20:42.19playdowill there be any progress in the near future?
20:42.29phhTS ?
20:42.50playdoyep
20:42.54playdowifi works
20:43.23Captnoordfor the hero ts framework to work it needs some rewrites
20:43.33Captnoordblackstone related
20:43.41phhshouldn't be too hard to do though
20:43.45playdoor doesnt the calib have something to do with the haptic
20:43.56phhno it has not :p
20:44.11playdok ^^
20:44.37playdobut it has changes during the weeks
20:44.42playdo-s +d
20:44.44phhadamw: any progress ?
20:45.01phhplaydo: what's your actual problem ?
20:45.33playdosome slight calibration issues i think ( with the middle button)
20:45.41Boydellphh  [17:54:28] <Boydell> man there are ALOT of warnings when compiling <--- I hate it too :( make a patch ! :p <------ I think I may start going through some of them and see whats up
20:45.43phhmiddle button ?!?
20:45.43adamwphh: nope, i'm getting nowhere - I just don't have the knowledge for this stuff, i am not a hacker let alone a hardware hacker. I just grep and read and guess and poke things. needs someone who understands things properly :/
20:46.03Boydelladamw: you sound like me :)
20:46.08adamwso I'm just doing what i'm actually supposed to do all day unless you or mrpippy can come up with anything :)
20:46.15playdoand responsiveness, e. g. my touch get recognised at the upper left although i press somewhere else
20:46.15phhadamw: that's not hardware problem :p
20:46.26adamwphh: well, I mean, a driver hacker.
20:46.31phhplaydo: what do you mean with middle button ?
20:46.38playdohmm
20:46.45playdomenu button
20:47.03phhadamw: don't worry, i'm not either. if one day some "driver hacker" sees my code, I think I'll hear me screaming from where I'm
20:47.06adamwphh: I can probably do anything you need to get any info off the phone that we turn out to need and I can build and test any changes you want to try out but I don't think I'm going to be able to actually fix it myself.
20:47.08adamwphh: haha :)
20:47.10phhplaydo: oh well might help then.
20:47.34adamwphh: I'm a qa/community guy, whatever little I can read and understand in actual code is purely from years of trial and error and reading changelogs, heh
20:47.44phhadamw: you have a "zoombar" right ?
20:47.52adamwthe keymaps stuff is at the level I can cope with, wifi drivers are not :)
20:47.53adamwphh: yes
20:48.01phhcan you see with getevent what happens to it ?
20:48.04adamwsure
20:48.05phhwith latest touchscreen driver
20:48.22stinebdspeaking of which
20:48.34phhstinebd: ?
20:48.42adamwphh: sure, gimme 5
20:48.53stinebdphh: tmzt expressed great displeasure with the current hacks we use for key layouts. is there any way to get this nonsense done in kernel? (ie raph110 vs raph100 keys)
20:49.15phhstinebd: I hate it kernel land, and I hate it user land.
20:49.21phhI have really no clue how to do that correctly
20:49.22makkonenhah
20:49.39stinebdphh: which part do you hate?
20:49.44phhstinebd: both ! :p
20:49.47makkonenc'mon, everyone... just EMBRACE the hackishness.
20:49.51phhI have no keyboard
20:49.53phhthat's *way* better
20:49.59phh(or not.)
20:50.06stinebdi'd just like the kernel to report proper keyboard for raph110
20:50.15phh(anyone want to offer me a TP or TP2 btw ?)
20:50.24stinebdi could steal my mother's rhod
20:50.30stinebdshe only uses it for phone calls and stuff
20:50.39phhI also use it only for "stuff" -_-'
20:50.56playdotry switching phones so she doesnt notice ^^
20:51.14stinebdan elaborate scheme
20:51.24stinebdi'll paint my old wizard black
20:51.32stinebdwith chrome sides
20:51.33playdofor example
20:51.36phh(batt and clock debug option at runtime pushed)
20:51.52playdoget some chrome sides from local car-tool-shop
20:52.01Boydellreplace it with a banada
20:52.04Boydell*banana
20:52.06adamwphh: I have the Stack O' Titans, but that's probably a bit old for you?
20:52.12Boydellse if she notices
20:52.15adamwno tp unfortunately and you're not having my tp2 :P
20:52.17phhthe what ?
20:52.32stinebdraph shouldn't be hard to find if you get several donations ;)
20:52.32adamwtitan is the CDMA version of the one before tp
20:52.35phhok
20:52.44adamwwhatever you call that, heh
20:52.48phhstinebd: I haven't got one for weeks.
20:53.01stinebddamn xda freeloaders
20:53.18phhI got a lot of donations during christmas though
20:53.22stinebdnotes he hasn't donated yet
20:53.30playdoxda freeloaders?
20:53.35adamwthere's some momentum in the tp2 thread to get enough donations to buy at least one tp2 for a dev, i'll definitely chip in for that
20:53.57adamwsplit $30 between a few people
20:53.59adamwthat's me covered
20:54.01phhTP2 already has devs -_-'
20:54.16adamwphh: who has a tp2 who actually knows what the hell they're doing?
20:54.21stinebdmakkonen
20:54.25adamwoh he does? didn't know
20:54.30phhstinebd: no
20:54.33Boydellhe does?
20:54.33phhhe has cdma raph
20:54.37stinebdoh
20:54.38Boydellhe has raph800?
20:54.46stinebdwho's doing all the rhod/topa commits?
20:54.51stinebdwas it mrpippy?
20:54.54phhadamw: hum. ok maybe not then.
20:54.57adamwmarkinus has a topaz
20:55.02adamw(I believe)
20:55.02phhstinebd: markinus
20:55.04stinebdahh
20:55.05phhadamw: yes
20:55.09adamwhe does most of the rhodium/topaz stuff
20:55.23stinebdall these m nicks
20:55.33emwehehe
20:55.38adamwhe doesn't seem to be able to do anything about the sound or wifi though - i know from an xda-dev topic that he tried with the sound, but didn't get anywhere...
20:55.56phhadamw: I don't think I could get sound working easily either ...
20:56.06playdoone other question: do u guys think its possible to learn to develop (maybe in c oder python) without having studied information tech?
20:56.17adamwwell two heads are always better than one
20:56.27adamwplaydo: yeah, I know lots of self-taught coders
20:56.35Boydellme :) ahha
20:56.42phhplaydo: I haven't ever written or read a single code-line at school yet.
20:56.44adamwdon't worry about boydell, there's some good ones too ;)
20:56.44playdoand what time does it take?
20:57.01Boydelladamw: hey now!
20:57.05playdophh: thanks, that makes me a lot more confident
20:57.24adamwphh: for the zoom strip info you need, does it matter if the ts isn't calibrated? I dunno if the zoom strip has anything to do with the touchscreen
20:57.27playdobut first i have to find the time
20:57.34phhadamw: it *is* the touchscreen :;p
20:57.38BoydellI think learning on your own is better. I had to learn C# in school (USELESS!) and i was so lost
20:57.49adamwphh: so i need to calibrate before getting you the info?
20:57.53playdo:-D thats like the keys of touch hd. ther "are" in the ts
20:57.56MrPippyi'll post my patches so you can try
20:58.01phhadamw: yes
20:58.12emwebtw, IIRC wiglleraway posted code he called a "zoombar driver" in the topaz android thread on xda.
20:58.13adamwoh fun, i haven't set up the new userspace yet. someone gimme the link again?
20:58.20phhplaydo: that's why you have to wait for new touchscreen driver :p
20:58.32playdoboydell: i learned a bit pascal in school. and something about decrypt and so on
20:58.46playdowants ts drivers
20:58.54playdowill code html for ts drivers
20:58.58MrPippyif i had a touch pro 2, i'd probably divide up the zoombar into new buttons since it has so few hardware ones
20:58.58Boydellschool is useless :\
20:59.06CaptnoordBoydell: amen to that
20:59.08playdobut it was fun ^^
20:59.16phhMrPippy: uh ? less than raph/diam ?
20:59.36stinebdit has four on the face
20:59.36emweMrPippy: hm. nice idea. i sometimes miss cursor keys ;)
20:59.37stinebdand power
20:59.47phhemwe: there is no cursor key ?!?
20:59.48stinebdand volume
20:59.50phhOUCH
20:59.54adamwphh: there are on the keyboard
20:59.57Boydellthe only thing a employer see's in a degree/deploma is that you know HOW to  learn. and your dedicated
20:59.57adamwbut not on the front of the device
21:00.00phhadamw: oh right.
21:00.04phhadamw: but topaz has none then ?
21:00.10adamwdunno about topaz, never saw one
21:00.11emwephh: topa100
21:00.42emweediting with android default keyboard is a pita without cursor keys
21:00.47playdoboydell: thats right, and you need about less than 20% in yout job from ur diploma
21:01.06adamwi still can't get the tp2 cursor keys mapped right
21:01.22Boydellagreed. Any job I've had required 0% so far lol
21:01.22adamwi dunno why, they don't seem to do the right thing when you map the correct eventcodes. oh well
21:02.02adamwphh: where's the new userspace? I lost the link
21:02.04playdohmm i need a few %, like knowing what a worm gear is and so on
21:02.20phhhttp://htcandroid.xland.cz/XDANDROID.02.02.10.7z
21:02.31adamwhas a degree in history and does linux qa for a living :P
21:02.32phhleobaillard: you should post it on the thread :p
21:02.32adamwphh: thanks
21:02.44Boydelladamw: you rock
21:02.57playdo+1
21:03.05adamwBoydell: except people keep asking me history questions and I don't remember a flipping thing, heh
21:03.14Captnoordhates boost
21:03.21Boydellhaha who does?
21:03.26adamwCaptnoord: oh god, everyone hates boost
21:03.34BoydellI was high in all my history classes in secondary school...
21:03.40adamwexcellent
21:03.55playdoi dropped history as soon as i could ^^
21:04.09Boydellit was required to graduate
21:04.12adamwi hated it in early high school, it's much more interesting later on and at degree level though.
21:04.18phhcan anyone port the usb otg driver ? :D
21:04.22BoydellI would imagine
21:04.50Captnoordthe only time I didn't like history was when we had a teacher who just didn't know how to tell stories..
21:05.34Captnoordphh what do you think of my desquire umull hack
21:05.34CaptnoordRunning The Desquirr decompiler plugin
21:05.34Captnoord-> Creating instruction list
21:05.34CaptnoordNo type information for function at 8005C708!
21:05.34CaptnoordNo type information for function at 8006CDD8!
21:05.35Captnoord80063008 - using conditional: 1
21:05.36Captnoord8006300C Warning: label without name
21:05.36CaptnoordNOTE: created new label loc_8006300C
21:05.37CaptnoordNo type information for function at 80062B94!
21:05.37Captnoordworps
21:05.38Captnoordcrap
21:05.42Captnoord800630d8   R3 = (R6 + (R6 << 2) << 1) * R2 >> 0x20 >> 3;
21:05.44Captnoordthat one
21:05.55phhok i'm a god, but not in asm or wince :p
21:06.01adamwpastebin!
21:06.08adamw(not that i'd understand it there either :>)
21:06.20phhadamw: he meant only to paste the last line :p
21:06.26adamwoh heh
21:06.27Captnoordnah umull is a u32 * u32 = u64 instruction
21:06.34phhCaptnoord: oO
21:06.36Captnoordbut only the high part is used
21:06.39Captnoordso I do
21:06.45Captnoordregister >> 32
21:06.46Captnoord:P
21:06.59Captnoordthat way its in 'normal' behavuare agai
21:07.04Captnoordresulting in that code
21:07.05Captnoordlolz
21:07.38phhcr2_: any clue of what needs to be done for usb otg ?
21:08.12makkonentouchscreen calibration is not working in that latest xdandroid package for anyone not on raph/diam. just a warning.
21:08.37phhmakkonen: oO
21:08.39adamwbtw, going back to keymaps, the cleanest design I can think of would be for all kernel drivers to return the same set of eventcodes, then we have a single android .kl for all devices which maps the eventcodes to an arbitrary set of keycodes, then we have as many .kcm.bins as we need for different variant devices to make the keys actually do what it says on the keycaps
21:09.12*** join/#htc-linux M1DLGpc (~M1DLGpc@bb-87-81-252-83.ukonline.co.uk)
21:09.54adamwshould be possible, might need a bit of design thinking based on the complete set of different keys found on all known variant hardware keyboards though
21:10.06MrPippyyeah i think thats what dream does, they have like 4 kcm/kl files in dream-open
21:10.21makkonenphh: little bug in the updated tssc-calibrate app, + the init script doesn't know what to do.
21:10.22adamwwell right now we have different kls *and* kcms for variants
21:10.34phhmakkonen: oh the input1 stuff, ok
21:10.38adamwit ought to be possible to have a single unified kl and handle all variations in kcms, i think
21:11.24adamwi could grab a bunch of screenshots and come up with a proof of concept I guess, I dunno. anyhoo. sorry phh, wiating on the userspace dl here
21:11.40phhadamw: makkonen is right, it won't work for you -_-'
21:11.43phhwell you can easily fix it
21:11.58adamwneed to poke something in rootfs/init ?
21:12.40phhyes
21:12.46phhs/input3/input1/
21:12.52makkonennot so easily. needs a new compile of tssc-calibrate
21:12.59adamwyep, input1 sounds right
21:13.15adamwmakkonen: i can probably do that, if you let me know where the fixed source is
21:13.16makkonenthe code that sends the 0 to calibration to close it never got updated to use the dynamically assigned device number.
21:13.26phhmakkonen: oops
21:13.55makkonen...or, for right now, you can just use someone else's calibration, and it should be close enough. just got emwe's to work.
21:14.27makkonenjust make a file, ts-calibration, in the root of the sd card, containing:
21:14.28makkonen153,915,868,921,510,548,155,186,866,136
21:14.37adamwmakkonen: well it's not about just making it work for me, it's for the touchstrip data phh wants, dunno how good the calibration needs to be for his purpsoes
21:14.46makkonenoh. fair enough.
21:15.13phhadamw: just correct would be enough
21:15.20adamwphh: would that work for you? or needs to be more precise?
21:15.23adamwok, i'll go with that then
21:15.44phhmakkonen: the ts-calibration file won't be written to the right place -_-'
21:15.54phhoh well if he also fixecs /dev/init.
21:15.57phh<PROTECTED>
21:16.04makkonenright. good point.
21:16.25adamwuh, /dev/init ? you mean the 'init' script in the top level of rootfs.img?
21:16.30adamwor something else?
21:16.40phhyes i meant /init
21:16.42makkonenadamw: I think I've got a working copy of tssc-calibrate, if you just want that.
21:16.43phhdon't know why I added a /dev there.
21:16.57adamwi don't know what i want any more :P
21:17.05makkonenhaha
21:17.23adamwwhat needs to go into /init , phh?
21:17.30phhadamw: s/input3/input1/
21:17.34adamwkk
21:17.35phhor maybe input2 now -_-'
21:17.36BoydellI want a million dollar, think we can work on the boys? lol
21:17.51phhBoydell: I prefer a bilion.
21:18.09makkonenok, so I've gotta do some trickery with sed to find the correct device number and pass that in the init script. that's next.
21:18.13adamwfinds a bloody 7zip extract tool
21:18.31Boydelleh. Then you have people bugging you, taking pictures and what not. Million sounds good :)
21:18.59phhBoydell: not enough to buy HTC and port android to TD.
21:19.09BoydellLMAO
21:19.23phhmakkonen: or you can return the dev id from the tssc-manager app
21:19.24phhand se it
21:19.25phhuse*
21:19.30adamwhell, go for a trillion, then you can buy google
21:19.30Boydellif you owned htc, you would no longer be using a TD
21:19.59phhBoydell: ok, a TP2.
21:20.05Boydellha
21:20.06*** join/#htc-linux Zoolooc_ (~fredsibar@p5495662C.dip.t-dialin.net)
21:20.06playdowhat about the temperature? is there any use of it atm?
21:20.11makkonenphh: hmm. I don't want to call that app if there's already a calibration file.
21:20.21Boydellor you can devolop something better ;)
21:20.25phhmakkonen: arf right.
21:20.32Boydelland i can be your "tester"
21:20.36makkonenno, this is good. I need to learn how to use sed.
21:20.44phhhum, TP2 with snapdragon
21:20.48phhand hdmi output
21:20.52Boydellgetting better
21:20.56makkonenthat would be so nice.
21:20.59Boydellbut isn the TP2 big
21:21.14makkonenI want a hd2 with a keyboard.
21:21.24phhmakkonen: that's what I said :p
21:21.30Boydelli want a nexus one with a keyboard :)
21:21.31makkonenbig is good. it's a pocket computer, how small does it need to be?
21:21.50makkonenphh: hd2's got about .6" more screen than the tp2.
21:21.54playdoit has to be smaller than my head, so it doesnt look stupid phoning with is
21:22.08Boydellplaydo: hahaahhaah
21:22.09playdo-s +t
21:22.30phhmakkonen: my TD is just the right size :p
21:22.42playdonah, size matters phh
21:22.50playdohd2 is my favorite too
21:22.52Boydellwell the diamond is thinner than the TP
21:22.53phhyes. it needs to be small enough.
21:23.18makkonenyeah, it does. I pick up people's iphones and think 'dammit. my screen has twice as many pixels... but this looks better, because it's so big. i hate you, iPhone.'
21:23.19phh(but a real size keyboard !)
21:23.36phh(easy to do no ? :D)
21:24.00playdoit just look better cause this fcuk capcitive ts
21:24.09adamwalright i think i hacked up my poor microsd card enough, let's boot it and see what happens :P
21:24.30playdois a dyslexic
21:24.39Boydellhey makkonen did you try and of the new commits for the main tree?
21:24.54adamwmakkonen: bah, sed is easy, do it with awk ;)
21:25.00Boydells/for/from
21:25.15MrPippymakkonen: if tssc-calibrate needs to know screen size, /init does some grep in /proc/cpuinfo, maybe call it from there?
21:25.24phhMrPippy: no it doesn't
21:25.25playdolast question before bedtime
21:25.32playdowhats the best irc client for winmo?
21:25.42adamwplaydo: they all suck, but zsirc is okay.
21:25.43phhMrPippy: it's only a device name problem (input1 vs input3)
21:25.57playdoadamw: thought that too as i messed through them
21:26.17playdotrying that one tomorrow on my tytn
21:26.20adamwplaydo: andchat alone is at least 95% of a good enough reason to prefer heavily broken android to working winmo :P
21:26.37playdoavailable free on market?
21:26.40adamwyeah
21:26.45MrPippyah
21:26.46playdothen i will give it a try tomorrow
21:27.08adamwhttp://www.happyassassin.net/extras/irc_meeting.jpg
21:27.10adamwthat's andchat
21:27.10playdoshould i use the new build with calib tool, or will it mess up my touch hd?
21:27.14makkonenboydell: I have not.
21:27.16*** part/#htc-linux TimY (~timy@rue92-7-88-164-189-67.fbx.proxad.net)
21:27.28phhplaydo: nothing will happens for you
21:27.40playdophh: thx
21:27.42Boydellok, when you get the chance, let me know how it goes. I think something there breaks my build.
21:28.31playdoadamw: wow, first client that looks awesome on first sight and not crappy winmo shit
21:28.56playdook, gotta try this now ^^
21:29.08adamwit's nice
21:29.09playdorushing for his class 2 sd card
21:29.14adamwit does multiserver and nick notifications as well
21:29.18adamwssl support
21:29.21adamwticks all my boxes
21:29.25adamwhum
21:29.36playdolets see how it looks with on screen kb......
21:29.43phhI need a quassel irc client.
21:30.13polyrhythmicAndChat rocks
21:30.15polyrhythmicI second that
21:30.36polyrhythmicalso ConnectBot beats the hell out of PocketPutty
21:30.52polyrhythmicWinMo is terrible for anything console-related
21:30.56adamwbah. having issues with the new rootfs. i think my keyboard map stuff broke it. :P
21:31.00playdotrue true
21:31.12MrPippyyeah connectbot is impressive
21:31.12adamwalso the calibration thing doesn't seem to have worked, i still have bad calibration. hold on while i poke stuff.
21:31.51Boydell*pokes adamw's phone
21:31.52playdobooting
21:32.12adamwoh wait, i think i screwed up, no surprises there!
21:32.53adamwshit, yeah, the tilt2 keymaps in the new xda env is broken
21:32.58adamwum, any xdandroid folks around?
21:33.55adamwoh well, for posterity, the permissions on tilt2_microp-keypad.kcm.bin are wrong, they're 711, should be 644. probably my fault somehow.
21:34.58playdozzzZZzzzZzZZzz
21:35.53playdohates first boot
21:36.01Boydellplaydo: agreed
21:36.23playdois it somewhat faster on internal storage?
21:36.24Boydellalso, No one has figured out how to open those .kcm.bin yet?
21:36.42adamwBoydell: well you can hexedit them, it's not that complex a layout
21:36.53adamwBoydell: but you may as well just make a new one and compile it
21:37.03adamwi don't think anyone's bothered to write a tool to decompile them yet
21:37.31Boydellwell my line of thought here was
21:38.09*** join/#htc-linux cr2_ (~cr2@ip-109-84-183-102.web.vodafone.de)
21:38.39playdowhat this failure: application android keyboard doesnt react?
21:38.56playdo(process com.android.inpumethod.latin)
21:39.01*** join/#htc-linux GregLem (~greg@lal69-5-88-171-98-250.fbx.proxad.net)
21:39.01adamwplaydo: just hit 'wait'
21:39.02phhignore it.
21:39.05adamwit doesn't mean much
21:39.26playdogot this quite often with several stuff. just ignored it like u said
21:39.57Boydellsorry
21:40.00Boydellok my thinking
21:40.05adamwphh: okay, booting up again, should have fixed it now i hope.
21:40.24Boydellthe /kl files maps key evernts to a "key" defined in the .kcm.bin?
21:40.24playdoquite touchy now
21:40.31playdomaybe just my imagination
21:40.46playdois happy
21:40.47adamwBoydell: they map events to a standard Android keycode, it has to be a keycode Android knows
21:40.54Boydellkey 51    COMMA
21:40.56adamwBoydell: you can then remap what the keycode actually does in the .kcm if you *want*
21:41.06adamwBoydell: yeah, look up KEYCODE_COMMA and you'll see it's in android
21:41.19adamwif you don't remap in the kcm file android just goes with whatever the default meaning is
21:41.42Boydellwell I'm just tring to get the capslock key working right
21:41.57Boydellbut i figured peering into the .kcm.bin would help
21:42.44playdo*stuck
21:42.56adamwBoydell: if you just want to see what a .kcm looks like, see http://www.happyassassin.net/extras/tilt2/tilt2_microp-keypad.kcm
21:43.19adamwBoydell: also read http://pdk.android.com/online-pdk/guide/keymaps_keyboard_input.html
21:43.37adamwwhy doesn't just hitting shift twice work for you, though?
21:43.42Boydellit does
21:43.52Boydellbut, we also have a CAPS button, that does nothing
21:44.06adamwi'd probably just map it to something else more useful then
21:44.28BoydellAh. Well I would like my keyboard to work like its suppose to
21:44.30stinebdyeah if anybody figures out how caps works, let me know ;)
21:44.57BoydellI've looked at the pdk too.
21:45.11*** join/#htc-linux leaigor (~laigor@188.134.16.241)
21:45.11stinebdandroid has it as a modifier but it just kills whichever kl it's in
21:45.49adamwstinebd: i didn't bother trying to figure it out since shift twice works fine anyway
21:45.51playdowifi works like a charm :-)
21:46.05phhit has been working fine for months.
21:46.14adamw...unless you have a broadcom :P
21:46.20phh:p
21:46.23stinebdit's when you want it to stop working that you have problems ;)
21:46.24playdonot according to some ppl on http://glemsom.anapnea.net/android/htc-msm-android/
21:46.48stinebddon't believe everything you read on there
21:47.14phh"don't ever trust users" is my motto.
21:47.20playdook :-)
21:47.24stinebdthat's like the tabloid news of kernel build services
21:47.49playdolike on dilbert: "Its Idiot proof" "If u cant use it, ure an idiot"
21:48.12adamwstinebd: boyd: you could try mapping caps lock key to some unused keycode and then mapping that keycode to the keysym for caps lock in the .kcm file
21:48.29*** join/#htc-linux Zoolooc__ (~fredsibar@p54954EC9.dip.t-dialin.net)
21:48.30Boydelllost me lol
21:48.31stinebdcaps lock isn't handled in .kcm
21:48.47stinebdthe modifiers are strictly handled in the layouts
21:48.56adamwstinebd: but I tried that trick with control and it doesn't work
21:49.07MrPippyemwe: still there?
21:49.15emweyap
21:49.22emwetesting out tssc_calibration ...
21:49.33adamwstinebd: i meant to try and make it *trigger* caps lock. but, yeah. i dunno. what keycode is supposed to trigger the caps mode? i can't find one
21:49.48stinebdadamw: do you mean which android keycode?
21:49.52adamwphh: ok, with that calibration data, my screen is inverted left/right
21:49.56adamwstinebd: yeah
21:50.00emweMrPippy: but not for thaaaat long anymore.
21:50.08stinebdadamw: hold on, let me grep the source to find out (pdk is outdated)
21:50.16phhadamw: inverts the data :p
21:50.23adamwphh: what should I swap?
21:50.40MrPippyok heres the patches http://opus.homelinux.com/public_files/bank2/
21:50.45*** join/#htc-linux M1DLGpc_ (~M1DLGpc@bb-87-81-252-83.ukonline.co.uk)
21:50.46stinebdadamw: CAPS_LOCK keycode flag
21:50.58stinebdadamw: note that's not an actual keycode itself, it's just a flag
21:51.15stinebdadamw: do you have a checkout of android source?
21:51.20adamwstinebd: right, my point is, what are you supposed to do with the caps_lock eventcode to make it trigger capslock
21:51.29adamwstinebd: not yet, life's been too short so far :P
21:51.36playdoargh crap, tochscreen stil got some issues. now its jumping to space on kb when pressing other keys
21:51.54MrPippyi'm also using two others that add separate workqueues for tty_io and smd_tty, i think those are for stability during heavy data
21:51.57Boydell<PROTECTED>
21:52.38emweMrPippy: wow, thanks! i can't believe i haven't seen any of the [PATCH x/y] where did you find them?! ;)
21:53.01emweMrPippy: that would have indicated to me earlier, that it's a series of patches ....
21:53.05MrPippyhehe i made them from my own tree, git format-patch
21:53.18stinebdadamw: (this is pure guessing, there is no proper documentation) in the .kl, i believe you're supposed to set the capslock key's scancode to an android keycode symbol (say SHIFT_LEFT for giggles) followed by the CAPS_LOCK flag (ie: key XX SHIFT_LEFT CAPS_LOCK)
21:53.26*** join/#htc-linux AndChat| (~AndChat@p57B3ADDD.dip.t-dialin.net)
21:53.27adamwMrPippy: this stuff is for accessing additional memory?
21:53.35MrPippyfirst two are modified from mainline kernel, third is the random one i found
21:53.40playdobb cu with andchat ^^
21:53.42stinebdadamw: i've never been able to get it to work. it hits that key in the .kl and dies, killing the rest of the .kl processing with it
21:53.46adamwstinebd: oh, so it's like the WAKE stuff
21:53.54stinebdadamw: right
21:54.01adamwfun
21:54.11bzohehe was impatient and patched my kernel manually for the 2nd banks tuff
21:54.14stinebdWAKE and WAKE_DROPPED are flags just like CAPS_LOCK
21:54.18adamwhave you ever tried to figure out ctrl? that's the other one I can't get
21:54.36adamwstinebd: maybe it's just broken in android? it happens. does any official android device have a dedicated capslock key?
21:54.39stinebdthere is no official android device with a caps lock key that i know of
21:54.48adamwcould just be broken then, they'd never know i guess
21:55.02AndChat|Cool its working
21:55.12MrPippyyeah i'm using extra ram and it seems stable
21:55.23AndChat|Kinda
21:55.23MrPippyandroid does caps lock by double-tap shift
21:55.39adamwMrPippy: yeah, we're trying to make a real caps lock key work though. which turns out tricky.
21:55.54adamwplaydo: nice
21:56.20playdoJust kb is a bit crappy
21:56.31adamwplaydo: you should get  an IRC proxy btw; i just typed this on a completely different systme to the last message :)
21:56.39playdoOr it is the ts calib
21:56.46emweMrPippy: eh? but you didn't add the "[PATCH x/y]" stuff to the subject lines, no?
21:57.12MrPippyformat-patch does it, i just told it the range of patches to dump
21:57.33*** join/#htc-linux playdo (~57b3addd@gateway/web/freenode/x-tgjjsbsnroptbgrr)
21:57.49playdodammit
21:58.08playdoapp shutdown
21:58.24BoydellMrPippy: is this bank2 stuff goin be pushed sometime soon :)
21:58.26playdobut nice experience
21:58.56adamwplaydo: yeah, it's decent. most important thing is it loads fast
21:59.09phhhas any TP2/TD2 user tried android's libgps ?
21:59.12adamwzsirc takes long enough to load that i could get my laptop out of my bag, open it up, set up tethering and get online with that faster :)
21:59.20playdoyeah, but its much better experience with a hw kb
21:59.36emweMrPippy: it's a pitty i won't be able to test this evening
21:59.51adamwphh: still getting calibration right btw - just booting with the values flipped
21:59.56playdojust port android to my tytn and i got a hardware kb :-)
22:00.17adamwMrPippy: are your patches specific to a particular device?
22:00.24Boydellplaydo: pretty sure there is aport going for that already..?
22:00.40playdohmm dont think so
22:00.41MrPippyno it should work with everything, although you'll have to enable the extra bank in your board- file
22:00.49adamwMrPippy: nice
22:01.03playdoboydell: wait, i take another short look at xda devs
22:01.09emweyah, congrats again ;)
22:01.31phhMrPippy: I hate you ! :p
22:01.35emwehehe
22:01.40phhnow I really need a new phone.
22:01.42emwelol
22:01.47emweyou see it works, phh
22:01.50emwedual bank ;)
22:01.56adamwgo on phh, get a rhodium, you know you want to ;)
22:02.01bzoMrPippy, is banks param passed as a kernel param, or should it be hardcoded in the board file?
22:02.10stinebdi know i want to ;)
22:02.12playdoboydell: there is something here http://wiki.xda-developers.com/index.php?pagename=Hermes_Linux
22:02.15phhadamw: yes I want to.
22:02.19playdobut im too tired now
22:02.21adamwphh: OK, got it up with okay calibration, lemme get you that zoom strip stuff
22:02.24phhadamw: I just don't really feel like paying one at their current price :p
22:02.42emwegnight everyone :/
22:02.55dcordeswho has a htc dream?
22:02.59Boydellplaydo: yah i remember seeing something, but i dont have that phone so meh
22:03.09MrPippyexcept for diamond, i think use_extra_bank=1 in cmdline should enable the 2nd bank
22:03.19playdoboydell: see ur point ;)
22:03.29playdois falling asleep
22:03.34playdogood night everyone
22:03.36stinebdis it stable?
22:04.13*** join/#htc-linux leobaillard (~leobailla@leobaillard.org)
22:05.20bzoMrPippy: my patches don't have the HOLES_IN_ZONE config in Kconfig, is that a remnant of another patch perhaps?
22:05.39adamwphh: what do you want exactly? want me to just redirect output to a file and then note down what i'm doing on it?
22:06.14phhadamw: what's the output form of getevent again ?
22:06.26phh<time> <type> <code> <value> ?
22:07.21adamwi don't know what all the values are
22:07.25adamwwhat i get for pressing 'e' is:
22:07.32adamwhttp://fpaste.org/yh97/
22:07.39adamwthat's a press and release, so three lines for each
22:07.39phhe ?
22:07.49adamwthat's just some output I have handy
22:07.49phhI was talking about zoom bar -_-'
22:08.03phhok so it's <type><code><value>
22:08.04adamwyeah i know but output's similar...
22:08.23phhbah just paste the result of getevent when playing with zoombar
22:08.25phhI'll filter it
22:09.14GNUtoohi,I'm a bit lost with all theses kernel arround, mickeyl would also like to have one kernel,I bet dcordes too....there are too much kernels arround....now we have most of the stuff working on leviathan's kernel but we lacks some features from codeaurora for xf86-video-msm....what should we do?
22:09.39bzowoohoo: free =  Mem:       182344       138748        43596
22:10.23adamwphh: hmm I'm not sure the calibration's good enough
22:10.35MrPippyyeah i guess HOLES_IN_ZONE is from another patch
22:10.40adamwphh: when I have the software keyboard up, swiping the zoombar rolls across the bottom line of the keyboard
22:10.58phhadamw: I'd say it's normal
22:11.04adamwoh, k then
22:11.06phhgive me the values, I'll say if it's or not :p
22:11.52dcordesGNUtoo, I second that. for the dream I clearly propose using latest codeaurora branch. a) because xf86-video-msm is also from codeaurora developers b) it seems to be the most recent. if you like we can import it in linuxtogo kernel.git
22:12.32MrPippyok i'm gonna push these
22:12.40stinebdprintf("%04x %04x %08x", event.type, event.code, event.value);
22:12.43GNUtoodcordes, leviathan has 2.6.32 from android
22:12.47stinebdphh: ^
22:12.52phhstinebd: ?
22:12.57stinebdthat's the output of getevent
22:13.08phhmmmmmuh ?
22:13.14bzo2nd bank patch is working well for me
22:13.16GNUtoodcordes, but the one from codeaurora has .31 which may be recent enough
22:13.25mickeylnot for my work
22:13.31mickeyli need .32
22:14.02MrPippyi've also got patches for smd_tty and tty_io to add separate workqueues, i think they're more stable but i've got to look at the smd_tty one more closely
22:14.05mickeylor we could port devtmpfs back
22:14.39GNUtoomickeyl, ok
22:14.52GNUtooah fast booting
22:15.12GNUtooso maybe we should try to port the msm fb things forward
22:15.30mickeylyep
22:15.33GNUtoowhat I fear is dri/drm (no 3d,just memory management)
22:15.38MrPippyplaying mp3s while running a dslr speed test...stable!
22:15.56mickeyli would actually prefer if we all could work on one kernel
22:16.04MrPippyi haven't tried my acid stability test yet...pandora
22:16.08mickeylbut it's your call
22:16.23bzonot that it's suprising, but 77mb of extra memory sure makes things a lot faster
22:17.15phhspeaking of that, my patch to low memory killer doesn't work.
22:17.24adamwphh: http://www.happyassassin.net/extras/zoombar2
22:17.39MrPippyyeah apps can stay running in the background and the disk cache is much bigger
22:17.59adamwphh: I swiped left-to-right twice, waited, swiped right to left twice, waited, touched left / middle / right then right / middle / left, waited, then left / middle / right right / middle / left again
22:18.01adamw(iirc)
22:18.17dcordesmickeyl, leviathan, GNUtoo you 3 are using the same machine with same aim. it would really make much sense to use one branch in the long run.
22:18.23phhadamw: uh ? i thaught zoombar was vertical ?
22:18.34MrPippyinteresting, nexus one just got an update that adds multitouch
22:18.42adamwphh: ? not on mine?
22:18.45stinebdzoombar goes across the bottom of the screen
22:18.50adamwyeah
22:18.53phhok
22:18.56mickeyldcordes: the reason for me to switch from linux-on-qualcomm to leviathan-incoming is the will to bring standard kernel interfaces
22:19.04adamwwell if you have it in landscape mode it's vertical i guess :)
22:19.10mickeylwhich helps my goal of not having to twist the middleware too much to adjust to all the androidisms
22:19.14adamwholding the phone portrait, the right way up, it's across the bottom of the screen
22:19.15GNUtoook
22:19.33*** join/#htc-linux GregLem (~greg@lal69-5-88-171-98-250.fbx.proxad.net)
22:19.35GNUtoowhat do you think of xf86-video-msm ?
22:19.50adamwphh: I guess it's event 0018, the others are probably me slipping or the bad calibration
22:19.55mickeylif it runs circles around kdrive, then lets go
22:20.05*** join/#htc-linux Markinus (~Miranda@gtng-4db043f4.pool.mediaWays.net)
22:20.05phhadamw: one is axis X the other is Y
22:20.10adamwah i see
22:20.18adamwso it's mostly horizontal heh
22:20.58adamwphh: can you give me a bit more info on the getevent output?
22:21.00*** join/#htc-linux zycho (~zycho@f054170061.adsl.alicedsl.de)
22:21.07adamwphh: i kept that letter 'e' output for a reason - here, lemme show you
22:21.08phhadamw: read /usr/inlucde/linux/input.h
22:21.34MarkinusHi
22:21.35phhgetevent output is <type><code><value>
22:21.36phhhi Markinus
22:21.42adamwookay, i'll try. but just in case you're feeling nice
22:21.43adamwhttp://fpaste.org/CS9Z/
22:22.03adamwthat's the 'e' letter on a tilt2 keyboard, and the 'push to talk' button under the volume buttons
22:22.16adamwthey give the exact same scancode - but that first line of output differs
22:22.16Markinusphh: did you see my gsensor patch?
22:22.19phhyes
22:22.40adamwso that's an event of type 4, whatever that is? hum.
22:22.49phhadamw: EV_MSC
22:23.09adamwMarkinus: is this helping at all? trying to figure out this 'letter e / ptt button' thing
22:23.28Markinusphh: But it isn'T working with 1.6 . . I don't know why . . maybe because it's compiled for 2.x . .
22:23.41phhMarkinus: hum, don't know
22:24.31phhMarkinus: guess what, you can commit to android-hardware too :p
22:24.35adamwMarkinus: so it seems that the letter 'e' and the ptt button give the exact same keyboard event but a differing EV_MSC event?
22:25.52Markinusphh: ok :)
22:26.10*** join/#htc-linux thedicemaster (thedicemas@j89126.upc-j.chello.nl)
22:27.31Markinusadamw: hmm, intresting, you have to many keys on your mobile! ;-)
22:27.37adamwMarkinus: :D
22:29.13adamwMarkinus: btw, just an idle q - have you tried to get the broadcom wireless (i'm assuming topaz has the same as rhodium) going at all? i poked it a bit with mrpippy's help but really didn't get anywhere
22:29.49Markinusadamw: Topaz has the same as Rhod/Dia/Blac etc.
22:30.18Markinusand it works :)
22:31.00adamwMarkinus: eh? it just makes the system reboot when i enable wireless on my tilt2
22:31.02MrPippydiam is tiwlan
22:31.05adamwsame for reefermattness
22:31.36*** join/#htc-linux zycho (~zycho@f054170061.adsl.alicedsl.de)
22:31.55Markinustilt2 has a other wlan chip . .
22:32.03adamwohh, i see...that's the difference
22:32.14adamwso tilt2 is different from tp2? huh.
22:32.19Markinusyeah
22:32.26adamwthat's new and annoying! :)
22:32.27Markinusin much thinks!
22:32.35adamwdidn't realize there was such a difference
22:32.38adamwwhat else is different?
22:32.39phhMarkinus: ?!? I thaught all tp2 had bcm4329
22:32.41phh27*
22:32.51adamwphh: i think you were right with 29
22:33.01phhadamw: no :p
22:33.10phhhum it's 25
22:33.17adamwphh: ooh. really?
22:33.18phh29 is for nexus
22:33.22phhwith fm receiver/transmitter
22:33.24Markinusphh: Topaz and Tilt2 has different chips or not?!?
22:33.28adamwphh: that is um interesting
22:33.30phhMarkinus: yes
22:33.37phhbut tilt2 is tp2 no ?
22:33.56Markinusyes
22:34.08phh[23:42:19] <adamw> so tilt2 is different from tp2? huh.
22:34.09phh[23:42:25] <Markinus> yeah
22:34.11adamwphh: hold on i'll show you what I mean
22:34.26phhadamw: 25seconds left.
22:34.29adamw:)
22:34.43adamwphh: briefly - look for the definition of SDIO_DEVICE_ID_BROADCOM_4325 in the source tree
22:34.47Markinusehhhhhhh
22:34.48adamwphh: it's defined to 0x000
22:34.52adamwer 0x0000
22:34.53phh(i've just found why lowmememorykiller doesn't work as expected and I hate that.)
22:34.58phhadamw: it's normal
22:35.02adamwoh kay
22:35.03phhI double checked it
22:35.10adamwcurses!
22:35.11MrPippyhmm this review says the topa has tiwlan, but i'm pretty sure rhod has bcm
22:35.16adamwyou're always one step ahead :)
22:35.28phhadamw: iirc you declare those id in the board file too
22:35.35phhMrPippy: yes
22:35.39phhtopa has ti, rhod has bcm
22:35.56Markinusyes. . . .
22:36.04adamwphh: um, really? I don't see anything for '4325' in board-htcrhodium.c
22:36.07MrPippyyay htc consistency!
22:36.12adamwsearching it for that string comes up blank
22:36.18phhadamw: it's in mmc.c
22:36.20adamwah
22:36.28phhand it's set to 0000 :p
22:36.35phhtime to sleep now.
22:36.39adamwphh: ok. :)
22:36.52*** part/#htc-linux c2d (~Mike@port-92-200-98-90.dynamic.qsc.de)
22:36.56adamwso the solution to my wireless woes is, bribe you with a tp2 and alcohol? :D
22:37.12Markinusha! ;-)
22:37.20stinebdthe tp2 is less important than the alcohol
22:38.19*** join/#htc-linux Zoolooc_ (~fredsibar@p549542D4.dip.t-dialin.net)
22:38.23adamwstinebd: hehe
22:38.28adamwthat's always the way
22:38.58*** join/#htc-linux GNUtoo|oeee (~GNUtoo@host46-130-dynamic.244-95-r.retail.telecomitalia.it)
22:40.45bzoMrPippy: hmm, had a reboot. Maybe because I was doing so many things at once since everything is faster?
22:41.04MrPippywas it downloading?
22:41.32bzoyeah, was in the market at the time
22:41.51bzologcat didn't report anything before it rebooted
22:43.23MrPippyi think these tty patches might help stability when data is involved
22:46.31bzohappened again in market. Though it wasn't a full reboot, it went directly to startup animation
22:48.56bzocellular data died and didn't come back with the semi-reboot
22:49.13MrPippyyeah lots of things get messed up after one of those, i would do a full reboot
22:49.34adamwmakkonen: do you have the fixed calibration tool handy? i'd like to get good calibration data for my phone done
22:51.07bzoheh, the performance is so much better, I'm almost willing to live with the instability :)
22:53.50*** join/#htc-linux Boydell (~Boydell@69.49.58.28)
22:54.32Boydelllive kernal messages is /proc/kmsg?
22:57.18Boydellsomeone is still here? :(
22:58.22MrPippyi don't know what kmsg is for
22:58.35MrPippyi just tailed it and it didn't seem to output anything
22:58.58*** join/#htc-linux kvaster (~kvaster@live.bn.by)
22:59.03bzoyes, cat it
22:59.04stinebd/proc/kmsg feeds dmesg and klogd
22:59.15Boydelli forgot cat in front
22:59.26Boydellcat /proc/kmsg
22:59.46adamwMrPippy: did you commit your bank2 thing yet?
22:59.50adamwwants to poke :)
23:00.04MrPippyyeah the patches are there
23:10.39*** join/#htc-linux GeekLad1 (~GeekLad@adsl-178-231-40.jax.bellsouth.net)
23:10.59*** join/#htc-linux GeekLad2 (~GeekLad@adsl-178-231-40.jax.bellsouth.net)
23:12.35adamwMrPippy: so i lost track - does the board- file need adjusting for this or not?
23:13.10adamwhmm, i see use_extra_bank in there already, so I guess not.
23:13.23MrPippyno but put use_extra_bank=1 in your cmdline
23:13.33adamwyup i caught htat.
23:24.51adamwMrPippy: hum, doesn't seem to work...still shows ~100MB of total memory, ~40MB free, nr_banks in dmesg is 1
23:26.09MrPippyoh thats right, the cmdline arg is "pmem.extra=1"
23:26.30adamwheh
23:26.34adamwokie!
23:27.24adamwhum, i already had that in there (no idea why)
23:27.37adamwmust have been cargo culted into the initial config i used
23:29.46adamwwould use_extra_bank=1 stop it working? if not, must be something else...
23:30.26MrPippymaybe its just "extra=1"?
23:30.36MrPippywhy do kernel params have to be so impossible to figure out
23:32.08adamwi'm looking at the same code now
23:32.21adamwhold on let's just look up what module_param_named does, heh
23:33.22MarkinusMrPippy: after the last commits the TS on the topaz doesn'T work anymore. I deactivate the TS Manager and the driver from the Topaz file, and the same . . Do you have a Idea what this is?
23:33.34adamwMrPippy: okay, plan B
23:33.42adamwMrPippy: i'll just flip the freakin' default and rebuild the kernel :D
23:35.04MrPippyi'm not sure markinus, are you giving it calibration points somehow?
23:35.37adamwMarkinus: tp2 is fine once I fed it some calibration data, fwiw
23:36.50MarkinusMrPippy: Did I need this when I deactivated the Manager driver?
23:37.44MrPippyfor the old driver i think the calibration was the msmts_calib cmdline argument
23:37.45MrPippyd
23:37.53MrPippydid you remove the warning from msm_ts.c (the old driver)?
23:39.10adamwMrPippy: ok, flipping the default value in the source did the trick.
23:39.21adamwMrPippy: 180MB total memory, nice :)
23:39.44adamwmax 178, free 84
23:40.45adamwholy crap that makes things faster
23:41.07MrPippywe could get more ram (maybe another 16MB) by reducing the pmem areas
23:41.07Markinusthere is a problem in the calibration, cannot create/find the the calibration points device or so (have to look one time mor for exact message, maybe is input3 something else) so I deactivated this for first to test my stuff. I use now the old defconfig with the calib parametrs, but there is no response from the tocuscreen . .
23:41.59bzoadamw: yes! I'm glad you see it too
23:42.23adamwbzo: launching marketplace was about 2x faster
23:42.52bzolaunching almost anything is 2x faster :)
23:42.57IceBoneOh, silly android.
23:43.42MarkinusMrPippy: Ohh, I see the warning now :-)
23:44.10Markinusyeah, its a reason  :P
23:44.54bzoMrPippy: any downside to reducing pmem areas?
23:46.04MrPippyandroid might want the space, but i don't know if the adsp area is used at all, and 32 MB for mdp is a lot
23:46.12MrPippyi should look at what nexus one has
23:47.11adamwdownloading a few things from marketplace hasn't made it reboot yet, pippy
23:48.35adamwexcellent stuff
23:48.48bzohave you tried any larger downlaods? astro file msr crashes every time
23:48.58adamwhow big's that?
23:49.05adamwi went with a ~5MB game
23:49.31*** join/#htc-linux jumoit (~samuel@121.77.87.145)
23:50.13bzobit less than 2mb
23:50.48adamwbzo: i got a checkout with the tty patches, maybe they do help
23:51.43bzoi just recompiled with the tty patch, trying again...
23:52.09MrPippyyeah i just grabbed astro and it was fine, going on 3.5 hours of uptime now
23:57.07bzook, worked this time

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