IRC log for #htc-linux on 20090113

00:01.58*** part/#htc-linux aaaaacccccc (n=acsvilup@151.67.31.27)
00:09.00*** join/#htc-linux konsta (n=asds@host81-154-247-76.range81-154.btcentralplus.com)
00:12.45*** join/#htc-linux exco (n=exco@e181067115.adsl.alicedsl.de)
00:17.30lupine_85readds the VDH to his kernel tree and rebuilds
00:17.49lupine_85ISTR someone saying that the "right" thing to do would be to write an X driver?
00:25.03lupine_85wonders if the problems with the touchscreen are being caused by ts.conf
00:28.47Untouchab1eGood night all
00:30.19*** join/#htc-linux t3chi3 (n=blarg@ip24-250-216-85.ga.at.cox.net)
00:32.57lupine_85yay, VDH "just worked" :)
00:38.14*** part/#htc-linux silven (n=zmc@64.209.10.12)
00:42.17*** join/#htc-linux addman3333 (n=skool@cpe-67-9-134-92.austin.res.rr.com)
00:57.52*** join/#htc-linux BHSPitLappy (n=BHSPitLa@unaffiliated/bhspitmonkey)
01:01.02*** part/#htc-linux exco (n=exco@e181067115.adsl.alicedsl.de)
01:02.19*** join/#htc-linux Tinyboom (n=nahh@108.84-49-166.nextgentel.com)
01:26.28*** join/#htc-linux zycho (n=zycho@a89-182-214-252.net-htp.de)
01:32.21*** join/#htc-linux exco (n=exco@e181067115.adsl.alicedsl.de)
01:32.34*** part/#htc-linux exco (n=exco@e181067115.adsl.alicedsl.de)
02:18.08lupine_8500:51 < lupine_85> OK, tslib is trying to use /dev/input/touchscreen0
02:18.08lupine_8500:51 < lupine_85> which doesn't exist
02:18.08lupine_8500:51 < lupine_85> setting it to event0 or mouse0 means X doesn't start
02:18.08lupine_8501:24 < lupine_85> mmhm, event0 et al aren't compatible with tslib
02:18.08lupine_8501:55 < lupine_85> unknown why
02:18.10lupine_8501:56 < lupine_85> X uses the 'mouse' driver in ps/2 mode instead, which gives the broken pointer controls
02:18.13lupine_8502:12 < lupine_85> (and -mouse mouse with vkeyb_toggle=off results in no screen update, whereas -mouse tslib with vkeyb_toggle=off gets screen updates...()
02:18.33lupine_85(I've been posting in the wrong channel :p)
02:19.45lupine_85j0b0, dcordes: I shall try recompiling tslib with debug on and against the latest kernel sources to make sure it's not something stupid. if that doesn't work, i'll dike out all the calibration and vkeyb code, and if *that* doesn't work, I'll see if we need a new driver in tslib
02:19.53tmztX -mouse tslib,,device=/dev/input/eventX
02:20.26tmztalso, evtest /dev/input/eventX and give me the top part of the output
02:20.43lupine_85is that different to symlinking event0 to touchscreen0 ?
02:21.22lupine_85(also, 'tis Xfbdev. though I guess it doesn't make a huge difference?)
02:21.31lupine_85is booting it back in now
02:23.13lupine_85tmzt: with that command-line, 'tis still "selected device is not a touchscreen I understand
02:23.13lupine_85"
02:23.13tmztyeah, those options only work with Xfbdev
02:23.14tmztgive me the evtest output (not the events)
02:23.48tmztcheck the driver for EV_BIT, ABS_BIT, input_set_abs_param
02:23.50lupine_85it's not in my current image
02:24.00lupine_85is compiling
02:24.27lupine_85we have input_report_abs(msm_ts_dev, ABS_X, x);
02:24.27lupine_85input_report_abs(msm_ts_dev, ABS_Y, y);
02:24.29lupine_85input_report_key(msm_ts_dev, BTN_TOUCH, 1);
02:24.47lupine_85in msm_ts.c
02:24.47lupine_85(I think that's the code path I'm following with vkeyb off)
02:24.50tmztset_param not report
02:25.37lupine_85whoops
02:25.37lupine_85input_set_abs_params(msm_ts_dev, ABS_X,
02:25.38lupine_85MSM_TS_ABS_X_MIN, MSM_TS_ABS_X_MAX, 0, 0);
02:25.38lupine_85input_set_abs_params(msm_ts_dev, ABS_Y,
02:25.38lupine_85MSM_TS_ABS_Y_MIN, MSM_TS_ABS_Y_MAX, 0, 0);
02:26.11tmztwhere's the driver source?
02:27.10*** join/#htc-linux nato2k (n=templarn@76.250.180.218)
02:27.17lupine_85drivers/input/touchscreen/msm_ts.c
02:31.43tmztwhich branch?
02:32.19*** join/#htc-linux reformatt (n=chatzill@71-37-165-208.bois.qwest.net)
02:32.31lupine_85htc-linux-2.6.25
02:32.36lupine_85http://rafb.net/p/mEbJlf18.html is the evtest output
02:33.01lupine_85(it all looks fine to me)
02:33.04*** join/#htc-linux br1ck_ (n=br1ck@xdslcq016.osnanet.de)
02:33.08tmztyeah
02:33.38lupine_85but at the moment, tslib won't take it
02:33.48tmztok, what does input_raw say?
02:33.51lupine_85and it's off-centre for -mouse mouse
02:33.55lupine_85input_raw ?
02:33.55tmzttslib_print_raw
02:34.31lupine_85that's a new one on me, I'm afraid
02:34.47*** join/#htc-linux woodson (n=CDP@c-76-101-90-149.hsd1.fl.comcast.net)
02:36.22tmztsorry, I can't remember what it's called but it's in tslib utils
02:36.44maejrephmm..  tmzt: know what's required to let a kernel module use an extern function that's defined in the kernel? :x
02:36.58tmztMODULE_EXPORT/MODULE_EXPORT_GPL
02:37.31maejrepin the providing code, or in the module?
02:37.38tmztthe providing module
02:37.59maejrepEXPORT_SYMBOL ?
02:38.00tmztis it in the core kernel or a module?
02:38.16tmztcheck another module
02:39.00maejrepthe function is compiled in, and I want to use it in the module
02:39.49maejrepgot it
02:39.51maejrepthanks
02:39.51tmztif it's in there core kernel I don't think you need to export it, what is happening?
02:41.25maejrepundefined symbol error on compiling the module
02:41.26lupine_85tmzt: ah, found the package, let's see if I can compile it
02:42.21maejrepworked tmzt
02:43.41maejrephmm, haret froze before starting the kernel :|
02:44.30lupine_85it'd help if ~I could find the source code...
02:45.13*** join/#htc-linux PoohbaLT (n=Poohba@c-98-235-66-242.hsd1.nj.comcast.net)
02:45.55maejreptrying again
02:46.11maejrepdamn :|  apparently I broke something
02:47.23lupine_85tmzt: I suspect the tools have been subsumed into tslib
02:47.25maejrepcr2: http://privatepaste.com/ffqrptEwuU -- new gpio_func patch with output enable
02:48.08tmztlupine_85: yeah, they build as a seperate package though
02:49.52lupine_85tmzt: no, I mean, the functionality has been subsumed into the source files so they're now a bunch of #ifdef'd printfs
02:55.49maejrephmm
02:56.27lupine_85tries to work out how to get bitbake to pass options to configure
02:56.56tmztedit .bb
02:57.11tmztor use bitbake devshell/devmode what ever
02:57.18*** join/#htc-linux piusvelte (n=irchon@71.175.8.29)
02:58.06lupine_85EXTRA_OECONF it would appear ?
02:59.30maejrepgot this sucks :|  i broke something and don't know what broke it D:
02:59.34maejreps/got/god/
02:59.52maejreptime to git reset :(
03:09.27maejrepNetRipper, yeah, kernel doesn't compile for me with git head
03:09.33maejrepundefined references to msm_proc_comm
03:12.33maejrephmm, nevermind, looks like my .config was broken :|
03:20.03*** join/#htc-linux _4nakin (n=Tuan@115.76.131.22)
03:25.03maejrepI now have no explanation for why git head won't boot on my phone anymore :(
03:34.32lupine_85tmzt: http://rafb.net/p/Z9Fe4x44.html is the check in tslib that's failing
03:35.31tmztI missed it in your last paste, no ABS_PRESSURE
03:35.39tmztadd that to the driver and it should work
03:39.22maejrepfound it, this time it was because I forgot to set my MTYPE back to 1910 after git reset :(
03:40.19maejreptmzt, so I'm setting alt-func 3 on my gpio, and same thing .. it fires once, but not again
03:42.13lupine_85tmzt: ooh, good point :)
03:42.19lupine_85gets to adding that
03:44.29lupine_85well. I asssume it'll be enough to just register set_abs_params for now?
03:46.29tmztyes, that macro sets the bit also
03:48.52lupine_85right, copy over & reboot time
03:49.46lupine_85by any chance, if we get a vkeyb and sensible calibration working in android and angstrom, can we dump the kernel-mode of those?
03:50.53lupine_85can't believe I didn't spot the pressure thing. I was comparing against another driver too :D
03:51.05lupine_85(except it didn't have pressue, and turns out, it has its own custom driver)
03:51.23tmztyeah, I going to look into making android device independent for input, I asked SanMehat about it but we'll have to look into the android input layer and see where to make the changes
03:52.45tmztI don't think vkey will end up being neccessary for kaiser/titan/raphael particulurly since android cupcake will have a virutal keyboard and input method layer
03:52.54tmztvirtual
03:53.05lupine_85plus raph has a ketyboard of awesome :D
03:53.23lupine_85it just really doesn't strike me as being appropriately kernel-mode
03:53.42tmztthe kernel calibration for touchscreen shouldn't be neccessary either, but I don't know if android will support raw data from a resistive touchscreen
03:53.55lupine_85it could probably be implemented as a user-space input driver? - ISTR that's possible?
03:54.17tmztusing uinput?
03:55.52tmztI thought that might be an approach for pre-cupcake virtual keyboard
03:56.36lupine_85alrighty! we get raw output
03:56.46lupine_85RAW---------------------> 288 339 0 238.110437
03:57.17tmztyeah, this is the same thing with dcordes's kaiser but that took a few weeks to find the problem
03:57.56nato2kso can i ask you guys a stupid qustion?  is there some kind of kaiser emulator out there that you are using or are you able to ssh into the phone once its booted into android/angstrom?  or are you just blindly compiling kernel's and booting them?
03:58.11lupine_85nato2k: ssh is our saviour
03:58.34nato2kgotcha, is there anywhere that i could find a tutorial for it?
03:58.47tmztnato2k: there's very little you can do until the kernel is up
03:58.57lupine_85tmzt: once I get angstrom working properly, it has a tempting option to try other peripheral drivers (like usb storae...)
03:59.00lupine_85storage*
03:59.05tmztnato2k: it appears future kernels will support console on usb serial
03:59.54tmztnato2k: there is no support for msm in qemu, even google uses a different architechure (called goldfish) for their android emulator
04:00.03lupine_85tmzt: when the vkeyb is "off", there's no X update either :/
04:00.21tmztlupine_85: yeah, do you have to video display "hack"?
04:00.23nato2kahh ok cool thanks for the info
04:00.26lupine_85tmzt: yeo
04:00.29lupine_85yep*
04:00.33lupine_85(the VDH)
04:00.41lupine_85works if msmvkeyb_toggle=hide
04:00.45nato2ki will leave you guys be.  its all way over my head, im a lowly windows software programmer :)  its impressive to watch you guys go at it.
04:00.53tmztwhere is the thread that updates the display, still in vkey
04:01.00tmztnato2k: do you have a kaiser?
04:01.01lupine_85i'd suspect so, yeah
04:01.04nato2kyes
04:01.34tmztdcordes has images at linuxtogo.org/~lgorris/
04:02.01nato2ktmzt: thanks.  ive been messing with it.
04:02.50tmztlupine_85: so it works?
04:02.51lupine_85actually, no - it's just an alternative setup of some variables in msm_fb.c
04:02.53lupine_85(the one I've got)
04:02.57nato2kunfortunately 1.65.24.36 runs like hell where i live so i keep a sd card to flash to it when im showing off to friends :)
04:03.24lupine_85tmzt: I've not been able to verify a moving cursor yet
04:03.28tmztah, that is the one with bad sd?
04:03.48lupine_85hopefully this boot...
04:03.57lupine_85vkeyb on
04:04.29lupine_85(and that makes psplash work too)
04:04.44tmztyeah, anything that uses the framebuffer should work with that
04:05.31lupine_85ho de hum, ts_calib still isn't working....
04:06.40tmztdon't use the xtscal, use the TSLIB_TSDEVICE=/dev/input/eventX tslib_calibrate
04:07.50maejrepnato2k, don't worry, I'm a web programmer by day ;p
04:08.39lupine_85tmzt: I have a ts_calibrate
04:08.46lupine_85no tslib_calibrate
04:09.01nato2kmaejrep: probably more exciting than what I do.  programming "power tools" for a health insurance company that ONLY supports visual basic /cinge
04:09.04lupine_85anyway, it gets me a calibration screen, but it doesn't register the clicks
04:09.43lupine_85hmm, is it missing the 'button' aspect?
04:10.24lupine_85we're doing BTN_TOUCH....
04:12.25tmztit might be the pressure is not reported
04:12.50tmztI don't think it uses BTN_TOUCH in the calibration, only records if there is pressure
04:12.56tmztbut I don't remember
04:13.21lupine_85ah
04:13.28lupine_85pressure is set at 0 all the time right now
04:13.36lupine_85put a pseudo event with pressure at 255 ?
04:13.42tmztjust report pressue 1 at the same time BTN_TOUCH
04:14.02lupine_85range is currently 0..255
04:14.10nato2kgood night all
04:14.40tmztI don't think it matters but I've always seen 0/1
04:14.57tmztand tslib doesn't care about the range either for some reason
04:15.01lupine_85okidoki
04:15.14lupine_85was going on the elo example
04:16.08lupine_85report pressure 0 0 BTN_TOUCH, 0) as well, I guess?
04:16.22tmztBTN_TOUCH 1
04:16.30tmzt0 when released
04:16.45lupine_85yep so pressure 1 when it's pushed down, pressure 0 when it's released
04:17.07tmztwhat's the line?
04:17.27lupine_85~msm_ts.c:196
04:17.37lupine_85there's an if(touch) {...} else {...}
04:17.40lupine_85touched*
04:17.56lupine_85the first {} is BTN_TOUCH, 1; the second BTN_TOUCH, 0
04:18.04lupine_85input_report_key(msm_ts_dev, BTN_TOUCH, 1);
04:18.10tmztright
04:18.24tmztoh, sorry
04:19.02lupine_85you know, just today I was reading about a kernel hotfix mechanism :p
04:19.02lupine_85sorry ?
04:19.13tmztI read what you said before incorrectly
04:19.25tmzt< lupine_85> report pressure 0 0 BTN_TOUCH, 0) as well, I guess?
04:20.33lupine_85eh, no worries. it's 4:20am after all :D
04:20.49lupine_85has an optician's appt at 9:20 and work after and girlfriend after that
04:24.46lupine_85haha
04:24.52lupine_85I think I just made it segfault
04:25.00lupine_85no, I haven't
04:25.58lupine_85woo
04:26.01lupine_85she works!
04:26.03lupine_85bitchin'
04:26.08lupine_85rawr
04:26.09lupine_85etc
04:26.56lupine_85angstrom also needs an xmodmap for the hardware keys, methinks
04:27.20tmztwhat keys are missing?
04:27.42lupine_85none that I know of, they just don't do anything yet
04:27.53tmztthe virtual keys?
04:28.13lupine_85hardware
04:28.16lupine_85the keypad
04:28.39tmztoh, can you run xev and find the keycodes?
04:28.48lupine_85/dev/input/event2
04:28.52lupine_85I could if it were in the image :D
04:28.59lupine_85I'll have to compile and add it... later
04:29.21lupine_85in the meantime, is the diff for the msm_ts.c handy for anyone? should I put it anywhere?
04:29.36tmztare you using -keybd evdev,,input= for that device?
04:29.53tmztgive it j0b0 I assume
04:30.49lupine_85I've not put a keybd stanza in that I know of
04:32.55maejrepi can't believe how frustrating this is :p
04:33.23lupine_85maejrep: feels so good when you get it though :D
04:33.29maejrepheh
04:34.10lupine_85j0b0: http://rafb.net/p/3fHpBx69.html
04:34.23lupine_85will post to the forum aussi
04:39.00_4nakinhi, i have android booted on my Himalaya, but the touchscreen doesn't work,
04:39.12_4nakinwhat should i do :(
04:40.42lupine_85tmzt: thanks again for the help :)
04:46.30tmzt_4nakin: does it work on any other filesystem?
04:46.59maejreplupine_85, did you ever find a way to enable those wince debug messages, or see where they get written to?
04:47.14lupine_85maejrep: I didn't particularly investigate, sorry
04:47.45lupine_85there is a log1: device...
04:49.01_4nakintmzt: wat do u mean ?
04:49.42lupine_85maejrep: I think, the anonymous sub_... routines that were before NKDebugPrintfW were actually asm implementations of the DEBUGMSG macro....
04:50.13maejrepwell its comparing bitmasks
04:50.27maejreplike "if this debug level flag is set, then print the message"
04:50.39lupine_85aye
04:50.50lupine_85I've not a clue how to enable that outputting, though
04:50.55maejrepwould be interesting to see that enabled though :p
04:51.11lupine_85maybe the #xda-devs people would know?
04:51.23maejrepPrints a string to the debug output stream. This function is called by the DEBUGMSG, RETAILMSG, and ERRORMSG macros when the macro requires output support.
04:55.06lupine_85it's not hugely helpful :/
04:55.51tmzt_4nakin: angstrom for instance, trying to debug this from just android would be very difficult
04:55.55_4nakintmzt: i've just tried on android a few days ago, i'm new on xda-dev forum
04:56.41tmzt_4nakin: it's important to know whether the kernel driver works, then probably integrate the kernel-based calibration/scaling like in dzo/j0b0 vkeyb
04:57.11_4nakintmzt:is there any tutorial ?
04:57.37tmztlupine_85: can you put the evdev and tslib ipks somewhere if they work on armv5?
04:58.37tmztI guess we'll have to see if dcordes can build those into an angstrom console image from armv5
04:58.53tmzts/from/for/
04:59.39_4nakintmzt:ok,thks, i will try on angstrom
05:01.10lupine_85tmzt: they're armv6 AFAIK
05:01.39lupine_85and the evtest is just an a.out :p
05:02.10lupine_85I'm happy to upload them if you want 'em though
05:02.31lupine_85has a whole dedicated server to work on this knotty problem :D
05:03.33tmztit would be good to have some test images for input issues
05:06.19*** join/#htc-linux daspsycho (n=Tim@HSI-KBW-082-212-005-059.hsi.kabelbw.de)
05:07.43lupine_85bah, framework-devel refuses to build
05:07.47lupine_85to fetch, even
05:10.26lupine_85tmzt: I plan on redoing the Angstrom part of the wiki sometime before I go to my optician's appointment today
05:11.23maejreplupine_85, know where I can find decent documentation on arm asm? :p
05:11.40lupine_85yep, I do :)
05:11.45lupine_85ibe second while I get you my links
05:12.21lupine_85http://www.simplemachines.it/doc/QRC0001H_rvct_v2.1_arm.pdf is your GOD
05:12.23lupine_85(http://infocenter.arm.com/help/topic/com.arm.doc.qrc0001l/QRC0001_UAL.pdf being the official link)
05:12.56lupine_85http://www.arm.com/miscPDFs/9658.pdf is also damn useful
05:13.36maejrepthanks :)
05:13.37lupine_85http://www.coranac.com/tonc/text/asm.htm#sec-arm is also OK
05:14.12lupine_85just remember to keep an eye on when Z is, and isn't, set, and you'll be fine
05:26.14*** join/#htc-linux goxboxlive (n=goxboxli@mail2.hjellnesconsult.no)
05:33.12*** join/#htc-linux woodson (n=CDP@c-76-101-90-149.hsd1.fl.comcast.net)
05:38.37*** join/#htc-linux nato2k (n=templarn@76.250.180.218)
05:52.53*** join/#htc-linux kiozen (n=oeichler@rgnb-5d87de48.pool.einsundeins.de)
06:03.43lupine_85installs bundles of stuff
06:10.17*** join/#htc-linux t3chi3 (n=blarg@ip24-250-216-85.ga.at.cox.net)
06:13.43*** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz)
06:24.30*** join/#htc-linux Tinyboom (n=nahh@108.84-49-166.nextgentel.com)
06:25.06*** part/#htc-linux _4nakin (n=Tuan@115.76.131.22)
06:30.34*** join/#htc-linux lpotter (n=ljp@58.173.176.153)
06:31.25*** join/#htc-linux zsircusr (n=zsirc@ip-77-24-211-128.web.vodafone.de)
06:41.35lupine_85http://www.lupine.me.uk/img/phone/ :)
06:45.03*** join/#htc-linux Moku (n=John@f048003122.adsl.alicedsl.de)
06:47.28zsircusrlupine_85 morning. nice screens :)
06:47.33zsircusrdoes ts work?
06:47.49lupine_85zsircusr: perfectly :)
06:48.00zsircusrwow
06:48.27lupine_85see the forum for details
06:48.27*** join/#htc-linux Mullins (n=bw@89.204.243.23)
06:48.34zsircusrcan you try fso-image ?
06:48.39zsircusrthe diam thread?
06:48.57lupine_85aye
06:49.32lupine_85and I'll build the fso image & have a go :)
06:49.57lupine_85basically, the touchscreen was missing pressure events
06:50.06lupine_85add those, tslib works like a charm
06:50.17dcordes-mid-of-ngood
06:50.18lupine_85hail tmzt for actually spotting it
06:50.39dcordes-mid-of-nhe fixed kais ts for x too
06:51.00lupine_85dcordes-mid-of-n: the development version of framworkd won't build here, incidentally
06:51.06lupine_85so it won't be the newest code
06:51.37dcordes-mid-of-nit won't work with the stable
06:51.56lupine_85it won't build with the newest
06:52.20tmzthey dcordes-mid-of-n , have you got the new framework-efl working?
06:52.31dcordes-mid-of-nframeworkd-devel
06:53.17lupine_85well. I'm still using org.openembedded.dev, but #PREFERRED_PROVIDER_frameworkd = "frameworkd-devel"
06:53.33lupine_85If I don't comment that out, it refuses to build - refuses to fetch, in fact
06:54.00dcordes-mid-of-nlupine_85 you need to build frameworkd-devel with the fso-:autorevs.inc
06:54.16lupine_85#require conf/distro/include/fso-autorev.inc
06:54.23lupine_85humm
06:54.32dcordes-mid-of-nhm
06:54.32lupine_85should that be before or after the other declaration?
06:54.48lupine_85or does it not matter?
06:55.05dcordes-mid-of-nguess the order doesn't matter
06:55.27dcordes-mid-of-ntmzt no I'm happy this place has inet :)
06:55.50lupine_85will give it another try
06:56.07lupine_85i'm assuming fso-image is x11?
06:57.14dcordes-mid-of-nI think so
06:57.21tmztlupine_85: if you want the real pressure data it might already be in the driver somewhere and you can use that instead of 0/1, tslib doesn't care as long as it can find a threashold in the average of the taps
06:58.32lupine_85tmzt: mm, it probably gets pulled with the rest of it, I just don't know which bit of data it actually is
06:58.46lupine_85and I'm not in an investigatory state right now :p
06:59.09dcordes-mid-of-ndoes it interfer with ts in android?
06:59.12tmztwhatever leads to the value for touched?? you mentioned earlier should be it
06:59.28lupine_85touched is basically boolean
06:59.35tmztdoes android do it's own calibration?
07:00.18tmztI mean what touched is derived from, the condition
07:00.33tmztit could just be in response to the irq but I don't think that's the case
07:00.38lupine_85touched = status & 0x100 ? 1 : 0;
07:00.44tmztoh
07:00.53lupine_85basically "is this keyup or keydown"?
07:00.57tmztI guess it is then, what is status?
07:01.07lupine_85status = readl(MSM_TS_BASE + TSSC_STATUS);
07:01.28lupine_85reg 0x10c
07:02.04tmztcould that be how stable the adc is? do we know that's really supposed to represent pen down?
07:02.16dcordes-mid-of-ndidb X need any hackery t draw correctly with the uptodate htc-msm-2.6.25 build?
07:02.38dcordes-mid-of-ns/build/code/
07:02.50lupine_85tmzt: couldn't say - I've not looked at the touchscreen code much at all
07:02.52lupine_85dcordes-mid-of-n: yes
07:03.17dcordes-mid-of-ngot a dif?
07:03.23lupine_85it required the Very Dirty Hack, and in addition, msmvkeyb_toggle can't be "of"
07:03.26lupine_85"off"
07:03.47tmztit seems the thread is not running when vkeyb isn't
07:03.57dcordes-mid-of-nsounds like the refresh issue
07:05.00lupine_85ERROR: Build of /opt/OE/openembedded/packages/linux/linux-msm7xxxx_git.bb do_setscene failed
07:05.22dcordes-mid-of-nnoo
07:05.44lupine_85dcordes-mid-of-n: full diff: http://rafb.net/p/d3EOQy55.html
07:06.01lupine_85ERROR: Exception:<type 'exceptions.AttributeError'> Message:'module' object has no attribute 'check_stamp_fn'
07:06.10tmztdcordes-mid-of-n: for future reference, what is the command to enter development mode on bitbake?
07:06.11lupine_85smells of typo ;)
07:06.56lupine_85dcordes-mid-of-n: is it much effort to actually go down the X driver route?
07:06.59dcordes-mid-of-ntmzt i don't use such mode
07:07.54dcordes-mid-of-nlupine_85 think tmzt knows better
07:08.02tmztlupine_85: I think we should get recovery.git working on raph, swetland_ said everything is there to draw to msmfb without the gpu
07:08.21tmzteverything except a couple of libraries in core.git I think
07:08.21lupine_85recovery.git?
07:08.26tmztandroid
07:08.36tmztpixelflinger
07:09.23lupine_85OK, and X can use that?
07:09.39*** join/#htc-linux ionstorm (n=ion@ip68-227-226-5.ph.ph.cox.net)
07:09.48tmztI don't know how to start writing an X driver for kdrive, I was waiting to see what would happen with X on arm devices, whether to go with the newest Xorg and a new kernel interface that would implement these things like kernel mode-setting and drm/dri on pc's
07:11.12tmztwhat happens here is the android msmfb is implemented like a drm based driver with off-screen memory but it's done through a single framebuffer which is larger then the displayed region
07:11.33dcordes-mid-of-nmaybe somebody coul start discussion in some X ml
07:11.39dcordes-mid-of-nd
07:12.25tmztthen one or more ioctl's are used to switch the visible screens, recovery.git does it one way and android 0.8 for qemu does it another way (UPDATE/PAN vs. changing the y-offset)
07:12.54tmztyeah, have we tried that or was that just about the touchscreen issue?
07:13.56tmztof course hw3d and libhgl work on raph??? but these rely on a library that's only on the g1's and I assume don't work on kaiser/titan,etc.
07:14.33*** join/#htc-linux gaintdot (n=moses@59.83.208.5)
07:19.23lupine_85libhgl /me assumes Mesa can be hacked into OpenGL|ES compliance?
07:20.28tmztlibagl is pretty close now and uses libhgl when necessary i think
07:21.18tmztI should say when available
07:22.03lupine_85I would love to have wesnoth on my phone :D
07:22.18tmztthat's sdl isn't it?
07:23.00lupine_85it pulled a bunch of gl libs when I started compiling it earlier
07:23.13lupine_85maybe sdl/gl by defaul
07:23.42tmztsdl loads gl through dlopen or something, they shouldn't be required
07:23.49lupine_85ho hum, time to repartition my SD card, I think
07:24.00lupine_85I might as well keep abreast of android as well :D
07:24.16BHSPitLappy"as well" x2!
07:25.04lupine_85:p
07:25.11lupine_85i'n tired
07:25.43tmztmaybe we could use Xnest or something on binder ipc/surface flinger, it could be simpler that a real ddx
07:26.10tmzttired as well, hope the information from maejre*p helps cr2 tommorrow
07:27.30*** join/#htc-linux marex (n=marex@vasut.kolej.mff.cuni.cz)
07:32.57*** join/#htc-linux pleemans (n=toi@116.54-246-81.adsl-static.isp.belgacom.be)
07:35.14*** join/#htc-linux rmoravcik (n=rmoravci@gtsgw.ttc.cz)
07:40.33*** join/#htc-linux StarLite (n=nnscript@s55916ca6.adsl.wanadoo.nl)
07:43.42*** join/#htc-linux daspsycho (n=Tim@HSI-KBW-082-212-005-059.hsi.kabelbw.de)
07:44.06*** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo)
07:55.41*** join/#htc-linux nebi (n=nebi@170.ftth2.cust.fyrobs1.upps.se.borderlight.net)
07:57.37lupine_85dcordes-mid-of-n: d'oh, I've beeen using the wrong bitbake
08:03.23dcordes-mid-of-noh
08:06.42lupine_85compile is running again now
08:07.19dcordes-mid-of-nok good
08:07.28lupine_85is generally fail at times :p
08:07.48dcordes-mid-of-nlow bat. ttyl
08:15.48*** join/#htc-linux pichurri (n=pichurri@users3.ilo.org)
08:17.14*** part/#htc-linux gaintdot (n=moses@59.83.208.5)
08:21.23lupine_85yay, halfway there
08:27.39NetRippertmzt, lupine_85, what is the ABS_PRESSURE patch for?
08:30.09lupine_85NetRipper: it brings /dev/input/event0 into compliance with tslib
08:31.07lupine_85(and I now kind of get what was being said earlier - the value of touched should == (or even .eql?) the value of pressure
08:32.07lupine_85NetRipper: combined with fugly msm_fb patch, it lets you run X and use the touchscreen
08:32.07lupine_85is verrr nice :)
08:32.20*** join/#htc-linux BabelO_ (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net)
08:32.23lupine_85drums his fingers at fso-image
08:32.56lupine_85is itching to tear the fb and vkeyb code out of the kernel
08:33.03NetRipperlupine_85, ah, ok, nice :)
08:33.31NetRipperthe vkeyb hack won't work anymore when the touchscreen interrupt is working
08:33.53NetRipperbut as we dont get it working it'll be fine for now.. when it does.. we'll fix it by adding a refresh thread (kaiser style)
08:34.10NetRipperbut the correct solution is to write an X msmfb driver
08:34.23lupine_85mmhm
08:34.40lupine_85waits for it to materialize :p
08:34.58lupine_85seems that the two main things that need to happen now are hardware keyboard, and onboard sound
08:35.09lupine_85(or at the very least, bluetooth :p)
08:35.18NetRipperthat PRESSURE thing doesn't break anything else does it?
08:35.27NetRipperi.e. android
08:35.38lupine_85NetRipper: haven't tested it, but can't see why it would
08:35.50lupine_85if android doesn't use the info, it should ignore it
08:35.50NetRipperok
08:35.59lupine_85if it does use the info, the numbers are sane
08:36.02NetRipperwill need to test it then, before committing to git ;)
08:36.15NetRipperok
08:36.15lupine_85has no commit powah :/
08:36.24NetRipperok
08:36.45lupine_85do we ACK patches or just bung them in as seems appropriate? ;)
08:36.58NetRipperi test them before i commit them
08:37.05NetRipperand try to sanatize them
08:37.09lupine_85ah, so it's mostly your branch?
08:37.27lupine_85(from an administrative point of view, that is)
08:37.33NetRipperwell, currently i commit a lot but the branch is not mine
08:37.47NetRipperit's just that the other devs dont have commit 'powah' either
08:37.56lupine_85hehe
08:38.08NetRipperthough they probably should create an account ;)
08:38.22lupine_85might have a stab at rebasing against latest master in the mid-term
08:38.38lupine_85oh, did you see screenshot? :) http://rafb.net/p/d3EOQy55.html
08:38.41lupine_85bah
08:38.46NetRipperyea, should be done as well sometime
08:38.48lupine_85http://www.lupine.me.uk/img/phone/Screenshot-1.png
08:39.07NetRipperlol, i saw the patch indeed ;)
08:39.07lupine_85is the general plan to keep it as a branch or get as much upstream as possible?
08:39.18lupine_85picture! thousand words! etc
08:39.21lupine_85:p
08:39.37NetRipperyou mean to get it included in the main kernel?
08:39.46NetRipperyea if the pic would load ;)
08:39.49lupine_85where appropriate, aye
08:39.57lupine_85humm, it loads here...
08:40.10NetRipperwell im not experienced enough with the kernel to do that cleanly, but it would be nice of course :)
08:40.25NetRipperwaiting for lupine.me.uk ;)
08:40.34lupine_85I have no kernel experience at all, beyond knowing that small diffs are good diffs :)
08:41.50lupine_85NetRipper: also, I have a dedicated host mostly aimed at this area, so expect services to start popping up over the next few weeks :p
08:41.57lupine_85(feel free to make requests)
08:42.23lupine_85might swapsies openembedded/angstrom as a mirror in exchange for them carrying the raph/diamond atch
08:42.27lupine_85arch*
08:42.43NetRipperaha
08:43.10NetRipperconnection timed out btw
08:43.16NetRipperi cant reach that host from here
08:43.44lupine_85http://forest.lupine.me.uk/gitweb/?p=kernel/.git;a=shortlog;h=htc-msm-2.6.25
08:43.47lupine_85ho de hum
08:43.52lupine_85what IP address is it resolving to?
08:44.00lupine_85(is it resolving at all?)
08:44.14lupine_85DNS might still be a bit wobbly as it only changed over yesterday
08:45.26NetRipperits not resolving
08:45.29NetRipperbut it works from my home
08:45.35NetRipperjust not from work
08:45.41NetRipperhttp://www.netripper.com/raphael/lupine1.jpg
08:45.43lupine_85it'll be a stale cache somewhere then
08:45.48NetRipperi mirrored it just to look at it ;p
08:46.00lupine_85heee
08:46.02lupine_85purdy, aye? :p
08:46.08NetRipperlooks nice :)
08:46.14lupine_85is of the opinion that 480x480 is the best size wallpaper
08:46.18NetRipperthat is angstrom ootb?
08:46.22lupine_85that way it fits regardless
08:46.25lupine_85well, the wallpaper is mine
08:46.38lupine_85the rest is pretty much OOTB (x11-image)
08:47.10NetRipperlooks nice, do you have the image online somewhere?
08:47.14*** join/#htc-linux woodyPL (i=woody@gateway/shell/blinkenshell.org/x-0d334264b31beb4c)
08:47.30NetRipperat least it looks more nerdy than android
08:47.41lupine_85in http://www.lupine.me.uk/img/random/ somewhere
08:47.44lupine_85nerdy? :p
08:47.59NetRippermore linux-like
08:48.29lupine_85ahh... yes, indeed it does
08:48.37lupine_85is ambivalent about android
08:48.46lupine_85on the one hand, it looks good, on the other hand, it's all.. yeah
08:48.55lupine_85doesn't like the dalvik thing
08:49.18lupine_85I do like the X server replacement thing
08:49.32lupine_85IMO, X is vastly over-specced for a single-user system
08:53.08lupine_85om nom nom pig sarnie for breakfast
08:53.14lupine_85(also vastly overspecced :D)
08:53.25lupine_85can't beat it for compatibility, of course
08:55.21NetRipperom no nom pig what?
08:56.53lupine_85sandwich
08:57.14NetRipperright, wasnt that easier to say? :p
08:57.22lupine_85nah :p
08:57.25*** join/#htc-linux _7lima_ (n=7lima@ham-dakosy-hsrp.eurogate.eu)
08:57.53lupine_85right, have fun without me - I expect everything to be working by tonight ;). /me -> eyetest -> eotk
08:57.57lupine_85work*
08:58.35NetRipperlol
08:58.41NetRipperi just started at work
08:58.50NetRipper;)
08:59.05NetRipperso you'll have to wait at least 8 hours before i take a look again ;)
09:03.00*** join/#htc-linux buster__ (n=buster@lostmind.org)
09:20.44*** join/#htc-linux Magister7 (n=Miranda@mail.vicont.in.ua)
09:34.16*** join/#htc-linux Babel1 (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net)
09:49.09*** join/#htc-linux radem205 (n=aaa@92-108-47-154.dynamic.upc.nl)
10:13.42*** join/#htc-linux rl2000 (n=rl2000@0x573b8b62.hhnqu1.dynamic.dsl.tele.dk)
10:19.38*** join/#htc-linux Untouchab1e (n=Untoucha@82.147.51.146)
10:19.46Untouchab1eHi all
10:21.57*** part/#htc-linux Untouchab1e (n=Untoucha@82.147.51.146)
10:22.02*** join/#htc-linux Untouchab1e (n=Untoucha@82.147.51.146)
10:26.03lupine_85NetRipper: meanie :p
10:26.10lupine_85is all eye-checked-up
10:26.44Untouchab1ehi lupine
10:26.50Untouchab1ehow are things?
10:26.56lupine_85I be tired
10:27.00lupine_85I'm also the only one in the office
10:27.09lupine_85so technically, I could nap
10:27.20Untouchab1ehehe
10:28.01Untouchab1eIm pretty tired as well, but napping is not an option.. Doing urban combat drills
10:29.57Untouchab1ejust have a nice break for lunch
10:34.55Skitzo__congrats on all the great work guys :)
10:35.22Untouchab1e:)
10:35.52Untouchab1eGlad you appreciate it.. (not that I am to be thanked).. Alot of great progress has been made over the last couple of weeks
10:36.50Skitzo__i think it was suggested in the thread on xda-dev, but if there is a way to donate to the people putting in all the hard dev work
10:37.37Untouchab1eIve offered my donations to more or less all of them, but I dont think they are doing it for the money ;)
10:37.54Untouchab1eI donated 35€ to Dccordes, but he offered to give it back ^^
10:38.01Skitzo__oh i know ;)
10:38.31Skitzo__i get the feeling most people are doing it out of hate for wm, or out of love for android or just for the fun of it
10:38.39lupine_85will always take money off of rubes^Wkind souls :p
10:38.41lupine_85nah, j/k
10:38.48lupine_85I've not actually done anything anyway :p
10:38.49Untouchab1eso now I contribute by hosting the latest builds for Android on the Raphael and helping out instead :)
10:39.25Untouchab1eI cant say I hate WinMo though, as Its still the OS that gives me the most possibilities with my phone. However, after getting the ADP1, i fell in love with Android
10:39.34Skitzo__ahhh
10:39.43lupine_85kicks WinMo while it's down
10:39.46Skitzo__has anyone tried contacting htc for device info?
10:39.46lupine_85kicks it again
10:40.10Untouchab1eI think WinMo 7 will be awesome though :)
10:40.34Untouchab1edont think HTC can help us out much Im afraid
10:40.43Skitzo__hrrm
10:42.23lupine_85pours diesel over it, sets it on fire, burns the ashes to cinders with Thermite, blocks the cinders and throws them into the ocean
10:42.29lupine_85*shrug* :p
10:47.11Untouchab1ehaha
10:47.26Untouchab1edont flame WinMo too much! It beats the shit out of everything else but Android
10:49.23NetRipperit's too easy to bash microsoft, so we do it even if we know it's not all that bad
10:49.23Skitzo__no way
10:50.12Skitzo__WinMo is just horribly designed
10:50.19Skitzo__from an end user point of view
10:51.14lupine_85NetRipper: I've avoided MS products like the plague since 2004
10:51.30lupine_85I just can't get on with them
10:51.47lupine_85other OSes being crap too doesn't mean we shouldn't flame WinMo ;)
10:52.10Skitzo__what desktop os do you run lupine_85 ?
10:52.14Skitzo__out of curiosity
10:52.23lupine_85Sometimes Debian, sometimes Ubuntu
10:52.31lupine_85I suspect I'll be going back to Debian shortly
10:52.44Skitzo__mm
10:52.53lupine_85Ubuntu's "Buy stuff / agree to licenses / do this / do that / this isn't broken honest!" attitude is grating again
10:53.15Untouchab1eWinMo just isnt designed for what mobiles phones have become today. For the old PDA's, it was perfect, and for mobile phones, although clunky, provides loads of advanced functionality not found in other OS's
10:53.20NetRipperi found the debian community to be too inactive around the time ubuntu got populr
10:53.43NetRipperdont know how it is now
10:53.52Untouchab1eTo be able to read and edit office documents, sync with Outlook, Wifi and all that on my old Qtek 8030 (or what ever it was called) was just insane at that time
10:54.09Skitzogranted Untouchab1e, but shouldn't MS keep up with the times?
10:54.12lupine_85mostly they sit around and moan about ubuntuers whilst ubuntuers ask for support :D
10:54.19lupine_85(on IRC, anyways)
10:54.29lupine_85still, Lenny is shaping up nicely
10:54.47Untouchab1eOff course, they have fallen behind.. but hopefully, Windows Mobile 7 will be a result of lessons learned from Google and Apple
10:54.57lupine_85we employ a debian packaging dude person here
10:55.02lupine_85he's rather cool
10:55.10Skitzodoesn't help all of us stuck with 6.1 though :P
10:56.08lupine_85fully intends to remove winmo from the phone as soon as $any_version_of linux is fully functional
10:56.19Skitzoagreed
10:56.22NetRipperat least the msm7201a devices have plenty of rom and ram to run windows 7 when it comes out ;)
10:56.39lupine_85take and make calls/sms, gives me an ssh client that isn't putty
10:56.45lupine_85it's not so much to ask :D
10:56.53Untouchab1eWindows Mobile 7 is supposedly more than compatible enough with phones such as the Raph and Diamond at least
10:57.16Untouchab1enot that we know much about WinMo 7 except that its going to use the camera to act like some kind of sensor..
10:57.19Untouchab1ebrb
10:58.16Untouchab1elunch over, better grab the rifle and head back out.. later all
10:58.38Skitzoseeya
11:30.32lupine_85grrs somewhat at fso-image
11:35.38*** join/#htc-linux Balsat (n=kll@87.72.13.87)
11:41.13lupine_85yay, past that hurdle
11:50.20lupine_851k tasks left... what's the worst that can happen? :p
11:58.07lupine_85attack of the .la files :p
11:58.18lupine_85presumably they all need stuffing into various -dbg packages?
12:31.48*** join/#htc-linux Guimli (n=guimli@ecu69-1-82-231-127-213.fbx.proxad.net)
12:36.02*** join/#htc-linux woodson (n=CDP@c-76-101-90-149.hsd1.fl.comcast.net)
12:40.07*** join/#htc-linux gmarzioz (n=gmarzioz@host11-34-dynamic.61-82-r.retail.telecomitalia.it)
12:40.27gmarziozhallo, everyone
12:40.47gmarziozcan you help me with Htcblueangel ??
12:50.10*** join/#htc-linux Guimli (n=guimli@ecu69-1-82-231-127-213.fbx.proxad.net)
12:59.28*** join/#htc-linux _7lima_ (n=7lima@ham-dakosy-hsrp.eurogate.eu)
13:10.53*** join/#htc-linux dcordes (n=zsirc@ip-77-24-132-241.web.vodafone.de)
13:11.17gmarziozcan you help me with Htcblueangel ??
13:11.28gmarziozhow to help??
13:20.51*** join/#htc-linux piusvelte (n=chatzill@nat.philau.edu)
13:21.53dcordesgmarzioz what is the problem?
13:23.23gmarziozthe battery not charge
13:23.32gmarziozthe wireless not work
13:23.45dcordesah that
13:24.35gmarziozi'm tring to complile kernel myself
13:24.46gmarziozis the right way ??
13:47.24*** join/#htc-linux Untouchab1e (n=Untoucha@82.147.51.146)
14:01.08gmarziozdcordes, no arch/arm/boot/zImage ......
14:01.30gmarziozafter make
14:04.54*** join/#htc-linux [1]Untouchab1e (n=Untoucha@82.147.51.146)
14:13.30BabelO_gmarzioz: what is your problem with battery charge ? how do you do ?
14:13.40BabelO_gmarzioz: and with wifi how do you do ?
14:13.53gmarziozsorry
14:14.00*** join/#htc-linux rmoravcik (n=rmoravci@gtsgw.ttc.cz)
14:14.18lupine_85eh? out of space?
14:14.33gmarziozBabelO_,i boot from haret
14:14.46BabelO_gmarzioz: and ?
14:14.59gmarziozwith angstrom
14:15.14gmarzioz2007-ipk-gclib
14:15.18gmarziozimage
14:15.28gmarziozand the battery not charge
14:15.41gmarziozthe red light go down
14:15.50BabelO_gmarzioz: and ? how do you charge it ? does your ac power is pluggeg or just USB ?
14:16.04gmarziozUSB
14:16.05BabelO_gmarzioz: if it is just usb, it can't work on linux, sorry
14:16.17BabelO_you have to plug ac power too
14:16.22gmarziozok
14:16.31gmarziozit's work in ac power
14:16.37BabelO_what about wifi ?
14:16.58gmarziozi don't find the access-point
14:17.25gmarziozthe ariel from gpe.2.8 not work
14:18.47gmarziozand after when i digit : ipkg update && upgrade the system don't boot later
14:18.50gmarziozwhy?
14:21.03gmarziozBabelO_, sorry
14:21.45BabelO_gmarzioz: try with iwlist wlan0 scan command on command line
14:21.48BabelO_or something like that
14:22.12gmarziozok
14:23.07AstainHellbringmorning
14:23.24*** join/#htc-linux Othello (i=Othello@gateway/tor/x-99c4f88446b28191)
14:24.15gmarziozBabelO_,but in winzoz the charge work
14:24.41gmarziozin USB
14:27.54gmarziozBabelO_, do you know?
14:30.30BabelO_gmarzioz: yes i know that, but the linux driver need to ask 500mA USB power, and default is just 100mA, so the linux driver miss some query for that
14:31.13gmarziozsorry
14:31.21gmarziozand for bluetooth
14:31.30gmarziozit not work
14:31.32BabelO_what for bluetooth ?
14:31.35BabelO_yes it works
14:31.49BabelO_hciattach command
14:31.53gmarziozok
14:31.56BabelO_on tty S3 idf i remember
14:32.00NetRipperhm i overheated my device up to the point where it turned itself off
14:32.16NetRipperby simply using the internet sharing winmo app
14:32.23gmarziozand for the ipkg upgrade
14:33.00BabelO_ipkg works if you have correct sourcelist
14:33.09BabelO_you need to do update first
14:33.15gmarziozyes
14:33.22gmarziozbut
14:33.25BabelO_i don't know anymore ipkg upgrade comment
14:33.27gmarziozafter reboot
14:33.32*** join/#htc-linux Nullbyte (i=nullbyte@85.17.58.168)
14:33.37gmarziozthe kernel not boot
14:33.42BabelO_?
14:33.58gmarziozyes
14:34.07BabelO_i do that several time on my device it works, then you have a bad server list where a bad kernel is available
14:34.11gmarziozthe boot stop at start up
14:34.26*** join/#htc-linux Guimli (n=guimli@ecu69-1-82-231-127-213.fbx.proxad.net)
14:34.42gmarziozwhere i can find the server list
14:34.44gmarzioz??
14:35.41*** join/#htc-linux PoohbaLT (n=Poohba@c-98-235-66-242.hsd1.nj.comcast.net)
14:38.51*** join/#htc-linux daspsycho_work (n=TimJorda@213.155.79.202)
14:40.45BabelO_gmarzioz: don't know
14:40.50BabelO_ask on #angstrom
14:40.55BabelO_if it exist
14:41.22BabelO_gmarzioz: or www.angstrom-distribution.org
14:41.27BabelO_shoulb be there
14:41.43gmarziozthanks
14:43.00lupine_85woohoo, I have an fso-image
14:43.10lupine_85wait, no I don't
14:43.12lupine_85it trickede me
14:53.43gmarziozBabelO_, sorry
14:53.55gmarziozwhat haret do ?
14:54.01gmarziozdo you know?
14:55.09gmarziozand why my sd not autorun when I plug in??
14:55.33gmarziozI do a vfat partition
14:56.00gmarziozand two subdirectory linux and 2577
14:56.12gmarziozi'm booting from wm6
14:57.06*** join/#htc-linux addman3333 (n=azachars@nat/sun/x-86487d8ada9fd9e2)
15:19.06*** join/#htc-linux exco (n=exco@e181084064.adsl.alicedsl.de)
15:19.57*** join/#htc-linux Guimli (n=guimli@ecu69-1-82-231-127-213.fbx.proxad.net)
15:22.32*** join/#htc-linux [1]Untouchab1e (n=Untoucha@82.147.51.146)
15:23.02*** join/#htc-linux GPFerror (n=gpferror@cpe-76-187-41-132.tx.res.rr.com)
16:00.31*** join/#htc-linux jaybird662 (n=jaybird6@ws168-75.icet.msstate.edu)
16:03.45*** join/#htc-linux _4nakin (n=Tuan@115.76.128.237)
16:04.12lupine_85Now I'm getting (I think) an angstrom tarball
16:04.13lupine_85woo, yay
16:05.11*** join/#htc-linux woodson (n=CDP@c-76-101-90-149.hsd1.fl.comcast.net)
16:07.13*** join/#htc-linux AstainHellbring (n=AstainHe@unaffiliated/astainhellbring)
16:12.18*** join/#htc-linux MethoS (n=clemens@host-091-097-244-164.ewe-ip-backbone.de)
16:13.20*** join/#htc-linux goxboxlive (n=goxboxli@24.84-48-212.nextgentel.com)
16:19.06*** join/#htc-linux Armand (n=Armand@SantaYnez-11-116.resnet.ucsb.edu)
16:19.58*** part/#htc-linux _4nakin (n=Tuan@115.76.128.237)
16:29.34*** part/#htc-linux jaybird662 (n=jaybird6@ws168-75.icet.msstate.edu)
17:00.41*** join/#htc-linux dcordes (n=dcordes@unaffiliated/dcordes)
17:05.43*** join/#htc-linux skola28 (n=skola28@32.148.19.159)
17:06.26dcordesNetRipper: I'm tethering using kaiser now and it doesn't heat at all
17:08.41skola28mine definately gets hot while tethered.  especialy via usb.
17:09.30dcordesskola28: your kais?
17:11.20dcordeslupine_85: that fso-image build worked?
17:11.40skola28yep... att tilt actually.
17:12.41dcordesI switched to display off mode
17:12.49dcordesI don't know if it affects clock
17:17.43*** join/#htc-linux pigeon (n=pigeon@60-241-137-179.static.tpgi.com.au)
17:19.58lupine_85dcordes: it built... at least
17:19.59lupine_85eventually
17:20.07lupine_85after working round $issues
17:20.07dcordesnice
17:20.12gmarzioz#ubuntu-it
17:20.14lupine_85I get to try it out later
17:20.28dcordescool let me know. I'll check logs
17:20.44dcordessomebody tried setting raphael audio routing?
17:25.46*** join/#htc-linux tsdogs (n=tsdogs@net70-17.metalit.net)
17:25.50*** join/#htc-linux Mullins07 (n=bw@89.204.251.249)
17:26.37*** join/#htc-linux SmeX (n=SmeX@ti121210a341-1298.bb.online.no)
17:30.20SmeXhey, does anyone in here have a Blackstone ?
17:30.27dcordeshi Mullins07
17:32.48SmeXno ?
17:33.43*** join/#htc-linux Rogro82 (n=rogro82@s5591104d.adsl.wanadoo.nl)
17:34.40Rogro82Dcordes: You said you used the vogue-smd, but somehow the smd1 isnt working on the latest commit. Did you also get the smd1 to work or only smd0?
17:38.44Rogro82dzo: Do you know why the smd1 isnt working at all on your latest commit? It is sending data but thats all..
17:39.21radem205I think dzo is sleeping :)
17:40.04dcordesRogro82: I did not try smd1. I used the htc-vogue git code for testing smd0
17:40.34dcordesSmeX: some people with blackstone frequent here who fight with the framebuffer
17:40.45Rogro82dcordes: k... smd0 is working for me, but using smd1 only writes but doesnt read any data
17:41.29dcordesdoes smd0 work stable for you with the git code?
17:43.28Rogro82dcordes: didnt try anything besides initializing for gprs... but its throwing creg and signal strength messages but only left it on for a few minutes
17:44.13dcordeswere you able to actually interact
17:44.24dcordesi.e. did it reply to your commands?
17:46.01Rogro82i think so... i only send the 2 initialization commands from setupmodem.sh
17:48.09Rogro82as far i can see the code the smd is using is pretty much the same as the one that was working (besides all the vogue stuff, but most of it isnt used when you compile it for a kaiser)... what happens when you try anything?
17:51.32dcordesbbl
17:52.14Rogro82is the check_for_smd_data a polling mechanism?
17:54.52*** join/#htc-linux cr2 (n=konversa@crpl6.physik.uni-wuppertal.de)
17:55.49Rogro82Anyone any knowledge of smd code? Is check_for_smd_data a polling mechanism which scans all channels in the smd channel list?
17:57.13*** part/#htc-linux BabelO_ (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net)
17:59.40cr2NetRipper: can you grep through the g1 tree for all vreg_ instances ?
18:01.46SmeXafaik the Omap linux kernel has suport for WVGA if that`s the problem we`re having with Linux on Blackstone
18:05.12*** join/#htc-linux bffn (n=chatzill@ua-85-227-18-133.cust.bredbandsbolaget.se)
18:06.57bffnAny news regarding the lcd issues on htc blackstone?
18:07.10*** join/#htc-linux RobSS (n=noone@rrcs-70-61-139-174.central.biz.rr.com)
18:13.50tmztcr2: you can do that with gitweb grep on android.git.kernel.org
18:14.55cr2tmzt: i don't have a linux machine connected to the net
18:15.33tmztI mean through a web browser
18:15.41tmztwill be back
18:18.12cr2omg. the offician g1 code :)
18:18.19cr2writel((1U << 16) | (~level & 0xffff),
18:18.19cr2<PROTECTED>
18:18.26cr2hahahaha
18:25.10*** join/#htc-linux imfloflo (n=imfloflo@cap31-6-88-180-73-121.fbx.proxad.net)
18:30.08*** join/#htc-linux timebomb (n=tb@e176121062.adsl.alicedsl.de)
18:31.25*** join/#htc-linux timebomb (n=tb@e176121062.adsl.alicedsl.de)
18:47.52*** join/#htc-linux reformatt (n=wmirc_us@174-144-25-240.pools.spcsdns.net)
18:49.44gmarziozi'm upgrading the kernelof my blueangel 2.6.21-hh20-r6 to 2.6.21-hh20-r16 it's all right
18:50.38*** join/#htc-linux reformatt (n=wmirc_us@174-144-25-240.pools.spcsdns.net)
18:58.05*** join/#htc-linux mightymouse (n=wmirc_us@174-144-25-240.pools.spcsdns.net)
18:59.10NetRippercr2, http://www.netripper.com/raphael/vregs.txt
19:02.21gmarziozhi
19:03.15*** join/#htc-linux Gnutoo (n=gnutoo@host13-24-dynamic.25-79-r.retail.telecomitalia.it)
19:05.19*** join/#htc-linux timebomb (n=tb@e176121062.adsl.alicedsl.de)
19:08.08*** join/#htc-linux mightymouse (n=wmirc_us@174-144-25-240.pools.spcsdns.net)
19:23.55*** join/#htc-linux goxboxlive (n=goxboxli@24.84-48-212.nextgentel.com)
19:28.28*** join/#htc-linux Balsat (n=kll@87.72.13.87)
19:30.09*** join/#htc-linux marmotta (n=skodde@unaffiliated/skodde)
19:31.20*** join/#htc-linux pleemans (n=toi@d54C2AAB7.access.telenet.be)
19:35.07StarLiteAndroid on HTC stuff sounds sooo nice
19:35.15StarLiteI wish my hermes could run it..
19:35.16StarLitebut well
19:35.32StarLiteguess my only chance will be my future X1 or TP :P
19:47.33*** join/#htc-linux marex (n=marex@vasut.kolej.mff.cuni.cz)
20:01.27*** join/#htc-linux kaze_ (n=kaze@pac33-1-82-235-251-34.fbx.proxad.net)
20:02.26kaze_hi there
20:02.33kaze_dzo do you happen to be around?
20:02.49*** join/#htc-linux Guimli1 (n=guimli@ecu69-1-82-231-127-213.fbx.proxad.net)
20:10.04*** join/#htc-linux BHSPitLappy (n=BHSPitLa@unaffiliated/bhspitmonkey)
20:10.18*** join/#htc-linux yoyey1 (n=yoann@lns-bzn-49f-81-56-185-87.adsl.proxad.net)
20:17.45*** join/#htc-linux Untouchab1e (n=Untoucha@82.147.51.146)
20:17.51*** part/#htc-linux Untouchab1e (n=Untoucha@82.147.51.146)
20:18.12*** join/#htc-linux Untouchab1e (n=Untoucha@82.147.51.146)
20:18.16Untouchab1egood evening all
20:20.17gmarziozhi
20:21.32Untouchab1ehow are things?
20:32.15*** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz)
20:32.35*** join/#htc-linux _7lima_ (n=7lima@p5099402d.dip0.t-ipconnect.de)
20:34.57*** join/#htc-linux timebomb (n=tb@e176121062.adsl.alicedsl.de)
20:36.07*** join/#htc-linux cr2 (n=cr2@ip-90-186-157-143.web.vodafone.de)
20:38.20cr2maejrep: vreg on/off is more or less resolved
20:38.56cr2don't yet know about vreg level
20:39.32*** join/#htc-linux MethoS- (n=lem@host-091-096-213-039.ewe-ip-backbone.de)
20:41.03*** join/#htc-linux SKyLiNe (n=SKyLiNe@195-240-0-202.ip.telfort.nl)
20:41.31[SKyLiNe]evening :)
20:42.02Untouchab1ehi
20:42.40tmztcr2: vdd?
20:42.53[SKyLiNe]can i ask a quick question? :)
20:43.00Untouchab1esure
20:43.09Untouchab1eask a away
20:43.11Untouchab1eaway*
20:43.51[SKyLiNe]great :D
20:44.18[SKyLiNe]ive just put android on my touch diamond, but i cant find the spare parts app to remap the end button of the virtual keyboard to go to the home screen :S :P
20:44.34cr2tmzt: multiple voltages
20:44.40Untouchab1ethats most likely because you dont have the cupcake build
20:44.46Untouchab1eof Android
20:44.51[SKyLiNe]true :)
20:45.00*** join/#htc-linux Zinbolic (n=zinbolic@0x57344a18.vgnxx3.dynamic.dsl.tele.dk)
20:45.10[SKyLiNe]got the build on the xda wiki
20:45.13*** join/#htc-linux Gluck (n=Gluck@av9592.comex.ru)
20:45.21Untouchab1eI host them though, but the packages on connect-utb.com are for the raph, but I can put together a Diamond package as well if you want
20:45.29cr2tmzt: they have created an unbelievable mess out of nothing
20:45.33[SKyLiNe]cupcake is the way to go? :)
20:45.46[SKyLiNe]whats the difference with the cupcake build? :)
20:45.53[SKyLiNe]hm would be nifty ;)
20:46.04cr2tmzt: and g1 still accesses the bkl pwm clk directly :)
20:47.16cr2tmzt: when will you buy a 7x0xA phone ? :)
20:47.17Untouchab1eSkyLiNE, if you go to connect-utb.com and download the oldest of the two packages for the Raphael..
20:47.31Untouchab1eextract the files from the downloaded .zip
20:48.00Untouchab1eand replace the file called... h/o
20:48.12tmztcr2: don't know unfortunately
20:48.44Untouchab1emy internet just came to a standstill..
20:49.00[SKyLiNe]replace it with what? ;)
20:49.21Untouchab1egive me a sec to look up the filename ^^
20:49.31cr2tmzt: they renamed vreg_BT vreg_GP6 which is the SD host power ;) isn't it a mess ?
20:49.33[SKyLiNe]haha ok no rush ;-)
20:49.43[SKyLiNe]brb too ill have a smoke ;)
20:49.59Untouchab1elol
20:50.05Untouchab1ethose will kill you soon :P
20:50.41tmztcr2: it looks like some of the headers in android kernel or from qualcomm, those are the ones that don't make much sense and are very different from the rest of linux and even the android team's code
20:51.33tmztcr2: I don't know what all the abstraction is about though, but we can't really expect all amss's to work the same can we? -)?
20:52.54*** join/#htc-linux MethoS- (n=lem@host-091-096-213-039.ewe-ip-backbone.de)
20:53.49Untouchab1ereplace the file initrd.droid.cpio.gz with the one in the package you downloaded from the XDA wiki
20:55.26*** join/#htc-linux Xime (n=xime@bankize.net)
20:55.32Untouchab1e:)
20:55.35Untouchab1egood luck
20:55.55[SKyLiNe]ah okay cheers mate :D will give that a try!
20:56.13[SKyLiNe]i also saw some cupcake.tgz on the xda forum but i dunno what it was build for
20:57.17j0b0Untouchab1e, how you adapt a set that works on raph to work on diam? whats the difference other than mtype in the haret file
20:57.27cr2tmzt: does somebody here have a sane g1 ?
20:57.36Untouchab1ehah, I have the ADP1
20:57.44Untouchab1eif thats considered sane
20:57.50cr2lol
20:57.55cr2exactly
20:58.12cr2can you read from /dev/mem ?
20:58.16tmztwasabi maybe
20:58.18Untouchab1ethink so..
20:58.21Untouchab1ebut not sure
20:58.32cr2i'm interested in the oemsbl
20:58.45*** join/#htc-linux timebomb (n=tb@e176121062.adsl.alicedsl.de)
20:58.53tmztI think cmonex has that
20:58.57tmztnot the device, the spl
20:58.58Untouchab1eah, unfortunately, I just got it, so havent been playing around with it so much yet
20:59.28cr2tmzt: i've started the amss disasm today. last time i've stopped it after 72h of cpu time spent ;)
20:59.47cr2tmzt: oemsbl != appsbl = spl
20:59.56tmztso radio?
21:00.07cr2radio = amss
21:00.23cr2g1 amss is actually interesting too
21:00.55tmztok, amss is the microkernel, oemsbl is the qualcomm arm9 bootloader?
21:00.58cr2but as i've said it is difficult to deal with
21:01.16cr2yes, some middle stage loader
21:01.33cr2there is QCSBL before it
21:01.37tmztviperbjk in #xda-devs might know something about it, he was looking for a way to access qcdiag on raph
21:02.19cr2why is qcdiag interesting ?
21:03.12tmzthis tool uses it to load small loaders/downloaders, don't know if it works on msm7x00 exactly or how far he's gotten with this
21:03.36tmztI think the g1 can only dump data through mtd so it's limited to what flash is accessible from arm11
21:04.02*** join/#htc-linux t3chi3 (n=blarg@n128-227-41-110.xlate.ufl.edu)
21:04.10*** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfddf9.pool.einsundeins.de)
21:04.28cr2i don't see how his tool can help me with my problem
21:05.45tmztis oemsbl in the radio image at all?
21:05.50cr2no
21:05.58cr2amss is the radio image
21:06.00tmztthen I'm not sure, sorry
21:10.45Untouchab1eany of you familiar with the qwerty.kcm?
21:23.05cr2NetRipper: are you here ?
21:34.29*** part/#htc-linux bffn (n=chatzill@ua-85-227-18-133.cust.bredbandsbolaget.se)
21:36.53*** join/#htc-linux bffn (n=jonas@ua-85-227-18-133.cust.bredbandsbolaget.se)
21:40.05tmzt17:25 < cmonex> instead of rtask  you do radiooptions 3 in android then use
21:40.08tmzt<PROTECTED>
21:40.11tmzt17:25 < cmonex> if cr2 needs an oemsbl, it's in any OTA or NBH, in radio.img
21:40.14tmzt17:25 < cmonex> no need to dump it..
21:40.16tmztcr2:
21:40.25NetRippercr2, off and on
21:40.34tmztthat's what I meant by radio.img, not the running image on arm9
21:41.08*** join/#htc-linux nebi (n=nebi@170.ftth2.cust.fyrobs1.upps.se.borderlight.net)
21:44.59cr2tmzt: i have an oemsbl from nbh
21:45.13cr2tmzt: what is ota ?
21:45.26tmztover the air, the update.zip's sent out my google/t-mobile
21:45.29tmztby
21:46.12cr2NetRipper: i have some vreg progress, so there is no excuse to non-working SD now. and wifi too
21:46.42cr2tmzt: ok, i'll finish my wiki docs first
21:52.37NetRippercr2, ok nice job..
21:52.44*** part/#htc-linux yoyey1 (n=yoann@lns-bzn-49f-81-56-185-87.adsl.proxad.net)
21:53.51NetRipperhow do we work its magic?
21:56.44cr2can you grep for vreg_ through your kernel source tree ?
21:57.11NetRipperi pasted the link
21:57.13NetRippersec
21:57.17NetRipperhttp://www.netripper.com/raphael/vregs.txt
21:59.02cr2ok, i've tracked all VREG_SWITCH options
21:59.39NetRippervreg_switch?
21:59.46cr2VREG_SET_LEVEL is something to be discussed with maejrep
21:59.49cr2yes
22:00.09NetRipperok, works differently than the current implementation?
22:00.12cr2it's done with the plain proc
22:00.21cr2_comm_wince masks
22:00.32NetRipperok
22:00.45NetRipperso it should work with the current proc_comm_wince implementation?
22:00.46cr2which is 2^g1idx
22:01.04cr2very smart from android people :)
22:01.06cr2yes
22:01.09NetRipperok
22:02.15*** join/#htc-linux SKyLiNe (n=SKyLiNe@195-240-0-202.ip.telfort.nl)
22:02.59SKyLiNehm virtual keyboard disabled by commandline.. is this new in the latest build or..
22:04.20cr2gp2,gp4,mmc,gp6 are used by g1
22:04.25cr2from your list
22:04.45NetRipperSKyLiNe, yes it is, use 'hide' when you just want to minimize it
22:04.52NetRipperor remove the option to make it work again
22:04.52cr2NetRipper: where is the board-raphael code ?
22:05.00NetRippercr2, sec
22:05.23NetRippercr2, http://git.linuxtogo.org/?p=groups/mobile-linux/kernel.git;a=tree;f=arch/arm/mach-msm;h=7499fe60271383e9a3baec8a83c18c7486c46458;hb=34c8afa173e6e8080eb2e26174e90ef0c29831e8
22:05.38SKyLiNehaha ok argh. ive got a touch diamond so no other input method then the vboard :P
22:06.02cr2arch/arm/mach-msm/board-htcraphael.c:vreg_mmc = vreg_get(0, "mmc");
22:06.07NetRipperSKyLiNe, the idea was that android (cupcake) has its own software keyboard now, making the virtual one unneccesary
22:06.47NetRippercr2, yes
22:07.52*** join/#htc-linux xota (n=xota@79.109.50.17.dyn.user.ono.com)
22:08.01SKyLiNeah okay cheers, but how can i input something on the touch diamond now then..
22:08.25xotahi! anybody with android over htc touch? is 100% compatible?
22:08.43NetRippercr2, that vregs[] list should be adjusted to your vregs on the vreg table on wiki i guess?
22:09.05cr2NetRipper: the sd vreg is 'gp6', look at the trout code
22:09.08NetRipperxota, i doubt 100%
22:09.45cr2NetRipper: i'd like to have an SD controller port usage table, depending on the device.
22:10.09xotaNetRipper: why?
22:10.11*** join/#htc-linux t3chi3 (n=blarg@n128-227-5-62.xlate.ufl.edu)
22:10.13cr2NetRipper: you'd also need to other vregs for LCD panel
22:10.28NetRippercr2, i dont see how you map gp6 to sd vreg
22:10.31cr2like gp2 and other strange names
22:11.05NetRipperxota, actually i dont know about progress on the touch
22:11.34cr2NetRipper: the vreg_switch api should be overridden, so it'll call proc_comm_wince/dex 0x15/0x16 calls
22:11.55NetRippercr2, yes but 'gp6' has id 23.. how do you map it to map it to your wiki page?
22:12.02NetRipperit has blidx 0 there
22:12.18cr22^g1idx for gp6 ?
22:12.43cr2i have not yet added the g1 table to wiki
22:13.11cr2gp6 should be proc_comm_wince/dex 0x800000
22:13.29cr2it's the "obsolete" BT vreg
22:14.23cr2the dex mask is 2^g1idx
22:14.25cr2i think
22:14.29NetRipperah
22:14.33NetRipperi see
22:15.00cr2not the blidx
22:15.19NetRipperwhy do you call it "bl"idx?
22:15.34cr2this stuff is rather convoluted, but i think i've got it straight now
22:15.44cr2because it's used in the spl as clock id
22:15.54NetRipperah, so bl from bootloader
22:15.56cr2check my raphLCD wiki page
22:16.46cr2NetRipper: can you write a table which SDx slot is used for what on all our A device ?
22:16.49NetRipperem, yes, im looking at it
22:17.01cr2raph*/diam*/bstone
22:17.33NetRipperok, i'll try.. i dont know all of them
22:17.59cr2start witht he data you know, i'll try to add the other
22:18.05NetRipperyep sure
22:18.09cr2g1 uses 'mmc' vreg for wifi
22:18.31cr2i think we will need some more weird vregs for it
22:18.43cr2with blidx c and d, i think
22:18.55cr2need to check the g1idx for them
22:19.09NetRippercr2, you mentioned yesterday or a few days back about being careful with the vregs, as it can fry (part of) the device
22:19.14cr2and our hitachi lcd needs 4,a,b
22:19.22cr2yes.
22:19.46cr2the raph500,800 may need other vregs enabled
22:19.51NetRipperis it like.. accidently using vreg for the wrong id.. or like.. can we actually set 5v on a device that only needs 2v?
22:20.01cr2that's why i'd like to have such info tables
22:20.10cr2not yet
22:20.22cr2we switch on only some preset voltages
22:20.27NetRipperok
22:20.34NetRippervoltages set by wince?
22:20.43cr2when we will implement the vreg_set_level we will need to be even more careful.
22:21.02cr2yes, or oemsbl et al.
22:21.30NetRipperso.. all existing calls to vreg_set_level should've be commented out already? ;)
22:21.36cr2that's why an accurate documentation is 90% of success
22:21.55NetRipperwe're just lucky it doesnt do any harm currently
22:22.09cr2the people who run raph100 code on 500/800 are really calling for trouble ;)
22:22.12cr2yeah
22:22.34NetRipperok
22:22.39NetRippernice to know
22:22.40NetRipper:D
22:22.49cr2the 500/800 people should document their lcd init like it's done for raph100
22:22.58NetRipperi didnt win the lottery but at least my raphael isnt fried ;)
22:23.18cr2and bstone is more like 500/800
22:23.51cr2that's they all have color palette troubles, sync ,pixlock and othe trouibles
22:24.15NetRipperyes
22:27.53cr2we should write 100% correct code for raph100, so others will be able to clone it
22:28.52NetRipperim currently stuck a bit on how im going to properly add proc_comm_wince to the kernel..
22:29.29NetRipperit's too bad it doesn't have the same interface as the existing proc_comm (as in, the constants differ)
22:29.46cr2please also remove the smc91x , serial0 and such junk from raph
22:29.53NetRipperwrites on todo
22:30.02cr2ifdef it with the amss version
22:30.18NetRipperproc_comm is amss specific?
22:30.22NetRipperhm
22:30.33cr2unless the androids will tell us how to query the amss version at runtime in a portable way
22:30.39cr2yes
22:30.51NetRipperso how do we call our amss
22:30.57cr2that's why we have _wince/dex after all
22:30.59NetRipperMSM_AMSS_VERSION_WINCE
22:31.22NetRipperbut still, i'll need to #ifdef on every proc_comm call.. and i wanted to avoid that ;)
22:31.24cr2something like that
22:31.47cr2it seems to be the same for all our deviceds
22:31.56cr2from vogue to raph
22:32.03NetRipperok
22:32.21cr2at least the api
22:32.52SKyLiNeim off to bed guys thanks for the help and keep up the good work ;-)
22:32.54NetRipperhm
22:33.01NetRipperso the whole ARM9 part is called the AMSS
22:33.05NetRippergood night SKyLiNe
22:33.36NetRipperthat's a better way than adding MSM_PROC_COMM_* config options ;)
22:33.40cr2amss is the modem/phone microkernel software that runs on arm9
22:33.49cr2yes
22:33.54NetRipperok
22:34.21xotawhat's the differences between HTC Touch and HTC Touch Vogue? this HTC: http://www.dmggrupo.com.pa/comercializadora/images/HTC%20TOUCH.jpg is always a Vogue? Or vogue is a version of this HTC?
22:34.55cr2vogue is touch in the NZ
22:35.04cr2or something like that
22:35.34cr2msm7500 based touch
22:35.37NetRippervogue is the CDMA version of touch
22:35.49xotacr2: NZ?
22:35.53NetRipperor rather
22:36.00cr2new zealand
22:36.13NetRipperHTC touch gsm is codenamed HTC elf, and HTC touch CDMA is codenamed HTC vogue
22:36.26*** join/#htc-linux Moku (n=John@f048003122.adsl.alicedsl.de)
22:37.03xotaoh! ok! CDMA version... I'm spanish, so... I thinks, that I need to wait for android ;D
22:37.06cr2NetRipper: don't use proc_comm in our code
22:37.32NetRippercr2, i know.. but i tried finding a more elegant solution than #ifdeffing everything ;)
22:37.40cr2362         /* A little vibrating welcome */
22:37.41cr2363         msm_proc_comm(PCOM_PMIC_VIBRA_ON, 0, 0);
22:37.41cr2use proc_comm_wince
22:38.00cr2or ifdef the proc_comm itself
22:38.02NetRippercr2, check the Makefile, that function is using the proc_comm_wince
22:38.18cr2and don't forget about the PCOM_ constants then
22:38.34NetRipperalso check the proc_comm.h.. it includes proc_comm_wince.h instead when raph/diam is detected..
22:38.36cr2NetRipper: it's confusing
22:38.37NetRipperbut its not very nice
22:38.50NetRipperexactly
22:39.08cr2since it's a function with different api
22:39.42cr2PCOM_GPIO_CFG is the next candidate for an override
22:40.03NetRipperim confused enough already, lol
22:40.03cr2but i need to understand the alt parameter mapping first
22:40.10cr2lol
22:40.34cr2the g1 proc_comm can't be used on our devices.
22:41.01NetRipperyes i understand.. did it like this to prevent #ifdefs all over the code..
22:41.03cr2because the androids have modified the amss to support an api different from wince
22:41.29cr2it prevents installing wince on g1 :)
22:41.34cr2which is a good thing
22:41.44cr2but also creates a lot of trouble for us
22:41.47NetRipperhehe
22:41.56NetRippergood thing for them is a bad thing for us
22:41.57NetRipper:)
22:42.18NetRipperi'll revert that proc_comm horror and make it 100% seperate
22:42.22cr2they broke the api in 3 areas
22:42.33cr2vreg control, gpio control and clk control
22:42.52cr2we need to override all 3
22:43.10cr2vreg_switch looks like the first candidate
22:43.13NetRipperthe gpio stuff seems to work pretty well already
22:43.38cr2not the PCOM_GPIO_CFG
22:43.54cr2which is the gpio_alt_func() config
22:44.04cr2you don't use it at all
22:44.08NetRipperah ok
22:44.20NetRippermaejrep made a new one for that
22:44.27cr2and we do not really document it on non-raph100 devices
22:44.27NetRipperhttp://privatepaste.com/ffqrptEwuU
22:44.41cr2yes, i know
22:44.44NetRipperok
22:44.55cr2but we need to use it in the drivers in appropriate places
22:45.10cr2and we need the docs for it
22:45.11NetRipperwhat is its use?
22:45.19NetRipperi.e. what does it do?
22:45.19cr2so far only raph100 is documented
22:45.46cr2arm9/arm11 owner, direction (IO) and function itself. irq config
22:46.13NetRipperok and can you give me an example of what you mean by "function" as i still fail to understand that part
22:46.34cr2PCOM_GPIO_CFG includes the pull current setting. maybe it's there too, i need to find out.
22:46.54cr2do you have raphael_gpio ?
22:47.36NetRipperon wiki
22:47.40NetRipperyes
22:47.45cr2hm. and we need to remove of ifdef all the weird android stuff from msm_ts.cv
22:47.53cr2ok.
22:48.48cr2look for gpio60
22:48.59NetRipperSCL
22:49.17cr260  0x3c    4,1  17  O  SCL
22:49.22cr2yes
22:49.22NetRipperyes
22:49.37cr2because it's configured as alt=4 function
22:50.03cr2you can disable alt=4, and it will be a "normal" gpio
22:50.17cr2you'll lose SCL then :)
22:50.30NetRipperaha
22:50.40cr2the problem is that there are not enough hw pins on the CPU
22:50.45cr2for all the functions
22:51.13cr2so there is some pin sharing mechanism
22:51.18NetRipperah
22:51.23cr2which is called 'alt func'
22:51.48cr2so it depends on the hardware config itself
22:51.59NetRipperright.. so '4' is a given
22:52.04cr2yes
22:52.05NetRipperit's the way it's wired
22:52.12NetRipperwhat is ,1 then?
22:52.20*** join/#htc-linux miknix (n=miknix@gentoo/developer/miknix)
22:52.21cr2that's why it's important to document FOOBAR_GPIO in wiki
22:52.39cr2because it's the hw info specific to that device
22:52.50cr2i've forgot what is 1 :)
22:52.58NetRipperlol
22:53.05cr2the wince gpio_alt api has 4 params
22:53.18cr2i think 1 is 'active high'
22:53.45NetRipperok as there's also some 0,7
22:54.05cr2a similar api exists for the pxa cpu, which has a good manual
22:54.08NetRipperor 0x48/0,0 ... o.O
22:54.20NetRipperbut ok
22:54.22NetRippersmall steps
22:54.25cr2<PROTECTED>
22:54.54cr2normally it's 0x48, but on sleep (or disabled)  it's 0
22:54.58NetRipperok, as in, when switched to the different alt, there's another device?
22:55.13NetRipperah
22:55.15cr2yes, 0x48 is some mixture that is written into the certain cpu reg
22:55.15NetRipperok
22:55.30cr2the PCOM_GPIO_CFG is more fine-grained
22:55.49cr2so i'd like to understand how do they correspond
22:56.00NetRipperso do we have any devices that share the same gpio that we know of?
22:56.06cr2then we can easily override PCOM_GPIO_CFG
22:56.25cr2yes
22:56.40cr2even raph100 and raph800 have different gpio alt config
22:56.54cr2SD card for example
22:57.07NetRipperyes i saw it..
22:57.35cr2that's why i'd like to have the SD controller usage table
22:57.53cr2and we need FOOBAR_GPIO page for each phone
22:58.24cr2even for the similar phones the gpio setup is different
22:58.33NetRipperbasically, so far i've seen.. all GSM phones have SD on SDC2 (base 1) and all CDMA phones have SD on SDC3 (base 1)
22:58.33cr2htc likes it this way
22:58.54cr2what about wifi ?
22:59.01NetRipperi dont know squad about wifi yet
22:59.13cr2ok
22:59.13NetRipperonly that it probably is on SDC1 for raph100
22:59.19cr2but we need this info too
22:59.29NetRipperas it's the only place i got a response besides SDC2
22:59.38cr2and we need some unified SDC naming
22:59.45NetRipper.. yes lol
22:59.53NetRipperbut it's mixed within g1 code too i think
22:59.56cr2because there are 3 SD* namings already
23:00.02NetRipperyes
23:00.10NetRipperSDIO, SDC and both base 0 and base 1
23:00.20cr2and SDC1_0
23:00.26cr2for the irq
23:01.13cr2SDC1_0 SDC1_1 SDC2_0 SDC2_1
23:01.19NetRipperyea
23:01.21NetRipperwe'll use that
23:01.30NetRipperbut in iomap
23:01.48cr2as long as it's unified, it does not matter
23:02.01cr2memorymap,gpio,irq
23:02.40NetRipperin iomap.h it's SDC1-4
23:03.39NetRipperand to make it worse.. the vogue also has SDC1-4 but has a physical offset of -0x100000
23:03.41*** join/#htc-linux dcordes (n=dcordes@unaffiliated/dcordes)
23:04.00cr2it's a non-A cpu
23:04.08NetRipperSDC1 starts at 0xa03* there where SDC1 starts at 0xa04* for us
23:04.09NetRipper;)
23:04.15cr2it has only 2 ports
23:04.29cr2well, it's a different cpu
23:04.32NetRipperok
23:04.36NetRipperbut same arch
23:04.42dcordeshi
23:04.45cr2the i2c clock divisor is at a different place too
23:04.50NetRipperhi dcordes
23:04.53cr2the irq controller is different
23:05.21cr2there are less gpios (bank5 and bank6)
23:05.24dcordescr2: you on the voda?
23:05.36cr2so the core cpu support is different
23:05.48NetRipperok
23:05.59cr2dcordes: they said t-mob umts does not work in my area. only edge
23:06.10dzoHi all, i fixed pio mode on the 7x00 sd controller, should help a bit.
23:06.13dcordesI'm in the westerwald area
23:06.21cr2dcordes: and eplus and o2 anyway
23:06.36dcordesfor a seminar
23:06.39NetRipperdzo, is it in linuxtogo git?
23:06.53dcordesthe only thing which works here is the vf hsda :)
23:07.01dcordeshi dzo
23:07.09dzoyes, haven't tried with new radio yet though.
23:07.30dcordesdzo: I'm running a very recent radio again
23:07.40dzohave you tried sd?
23:07.53cr2dzo: does it support 48MHz for 2.0 cards ?
23:07.58dcordesonly with the commit from the other day
23:08.02NetRipperwow, quite a patch
23:08.04dcordesand with the 1.6 radio
23:08.15dcordesdzo: I have posted my results in the kaiser thread on xda
23:08.32dzocr2: max speed i can get is 32MHz
23:09.04dcordesdzo: with your initial pio code and the 1.6 radio, SD didn't work at all for me
23:09.25cr2dzo: sorry, i forgot that you have a non-A cpu
23:09.41dcordescr2: what's 2.0 card?
23:10.04dzolatest patch works for me, i had to back to the polling driver, seems pio interrupts are stuffed.
23:10.49NetRipperim off to bed, good night guys
23:10.53cr2dcordes: SD 2.0 spec
23:10.55dcordesnn NetRipper
23:11.01cr2good night
23:11.55cr2dcordes: i got an nc10 with the flatrate umts. running windows for now
23:12.12dcordesin this rural area I'm currently located in, the connection latency is way lower at night time
23:12.51dcordescr2: I'm using tethering with kaiser->laptop
23:13.13cr2dcordes: the umts modem is an usb device here
23:13.19dcordesin the afternoon it was so slow.. now it's like my home dsl connection
23:13.30*** part/#htc-linux Nullbyte (i=nullbyte@85.17.58.168)
23:13.32cr2dcordes: i may check the raph
23:13.55cr2i think i'm the only user in this area :) so i have full speed.
23:14.16*** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo)
23:14.38dcordesNetRipper said it likes to overheat when used for tethering
23:14.42dcordesha right now it seems the same for me
23:14.46dcordeshi BabelO
23:39.34*** join/#htc-linux t3chi3 (n=blarg@ip24-250-216-85.ga.at.cox.net)
23:43.17*** part/#htc-linux Balsat (n=kll@87.72.13.87)
23:56.22NAiLAnyone that can recommend a good firmware for my Kaiser? :)

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