IRC log for #htc-linux on 20090107

00:13.50excoimfloflo, Rogro82: how long do you estimate until we can fully use Android on our TC's?
00:22.39*** join/#htc-linux Tinyboom (n=nahh@108.84-49-166.nextgentel.com)
00:26.25*** join/#htc-linux Mullins (n=bw@89.204.244.30)
00:27.51lupine_85looks at building an angstrom image overnight
00:30.51*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
00:40.19NetRipperj0b0, are you working on that stuff in board-trout-keypad.c? the EV_SW/SW_LID?
00:56.07imflofloexco dunno depends on DZO progress for the moment
01:04.20*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
01:11.38j0b0NetRipper i was earlier but no breakthrough yet
01:12.16*** join/#htc-linux bachflomid (n=bachflom@70-59-66-3.mpls.qwest.net)
01:12.38j0b0also i think android assumes keypad input on /dev/event0 at least in places
01:12.46j0b0where we have the touchscreen
01:14.54j0b0#ifdef HAVE_ANDROID_OS
01:14.54j0b0<PROTECTED>
01:14.54j0b0<PROTECTED>
01:16.14NetRipperok but touchscreen is input0
01:16.18NetRippernot event, i think
01:17.49tmztit's both
01:18.03tmztcat /sys/class/input/*/name
01:18.23j0b0it is class/input/input0 but /dev/input/event0  when i cat it and stroke the screen i see stuff spewed
01:19.51NetRipperaha
01:23.34*** join/#htc-linux Rogro82 (n=rogro82@s5591104d.adsl.wanadoo.nl)
01:31.57j0b0i must sleep now tho
01:33.56NetRipperyea me too, but this stuff is too much fun lol
01:34.00NetRippergood night though;)
01:34.12j0b0truste
01:34.22*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
01:38.41excogood night channel
01:39.18*** part/#htc-linux exco (n=exco@e181076145.adsl.alicedsl.de)
01:41.51*** join/#htc-linux Heema (n=weechat@217.139.57.79)
01:42.45tmztNetRipper: so this needs changed on the android side?
01:43.38NetRipperwell, would be nice if they read multiple devices, if android can't we should make a hack in the kernel that combines events from multiple devices i guess?
01:44.19NetRipperbut currently android also has touchscreen and keypad split, so i doubt there's a problem..
01:48.30*** join/#htc-linux jerry507_mac (n=jerry507@jsmac.student.iastate.edu)
01:54.22lupine_85heh. turns out that compiles go faster when you throw an additional 3,200MHz at them
01:54.30lupine_85stupid powernowd
02:02.07NetRipper:p
02:02.28NetRipperand add -j<nr_of_cores+1> to the make command
02:02.28NetRipper;)
02:03.07NetRipperim off as well, good night
02:03.47*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
02:10.51*** part/#htc-linux Heema (n=weechat@217.139.57.79)
02:18.53*** join/#htc-linux curiousblue (n=curiousb@gates2.resnet.susx.ac.uk)
02:19.03curiousbluehi guys
02:19.09curiousbluewhats up with xda-developes?
02:19.13curiousbluedevelopers*
02:29.48imfloflodown
02:32.34*** join/#htc-linux DracoDan (n=F@pool-71-163-138-224.washdc.east.verizon.net)
02:33.49*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
02:35.02*** join/#htc-linux br1ck (n=br1ck@xdslcy254.osnanet.de)
02:50.47DracoDanI'm intrueged by this project, how can i see whther there are any plans to work on my phone? (verizon HTC touch pro)
02:51.58tmztDracoDan: raph500? there has been some work but it appears it's very different from the other Raphael devices
02:51.59*** join/#htc-linux Mullins (n=bw@89.204.244.30)
02:53.22maejreptmzt, its very different from raph800 as well?
02:54.03DracoDanyeah, i thought the only differences were the amount of memory, keyboard, and radio
02:54.59tmztmaejrep: yeah, it's been discussed here that it could be msm7500 based rather than msm7501a
02:55.11tmztand has less ram
02:55.11maejrepah
02:55.21tmztDracoDan: can you run haret on the device?
02:55.31DracoDangod im starting to hate verizon....
02:55.47DracoDanim not familiar with "haret"
02:56.02tmzthandhelds.org wiki HaRET
02:56.17tmzthandhelds.org/~koconnor/haret/
02:56.36tmztget the latest version and haretconsole with the same date
02:56.52maejrepDracoDan, it'll be the best way to contribute for supporting your device
02:57.04tmztyou can use haretconsole with python or just use telnet
02:57.14tmztmaejrep: I think parmaster has one now
02:57.22maejrepah
02:58.29tmztmaejrep: what cpu is reported for raph800 in haret?
02:58.58maejrepisn't that just built into haret?
02:59.30maejrepWelcome, this is HaRET pre-0.5.3-20081029_002018 running on WindowsCE v5.2
02:59.30maejrepMinimal virtual address: 00010000, maximal virtual address: 7FFFFFFF
02:59.30maejrepDetected machine Raphael/MSM7xxxA (Plat='PocketPC' OEM='RAPH800')
02:59.30maejrepCPU is ARM ARM arch 6 stepping 2 running in system mode
02:59.50tmztoh
03:00.06tmztok, what does windows report in control panel system?
03:00.18tmzt(I don't have a ppc so not sure where)
03:00.50maejrepwindows host, or WM?
03:00.56tmztwm
03:01.07maejrepDevice Info says "QUALCOMM(R) 7501A" :)
03:01.19tmztah
03:01.28tmztDracoDan: what does yours say?
03:01.51maejrepbrb
03:03.28DracoDanhad to hunt down my cable
03:03.35DracoDanor (a) cable
03:03.50DracoDaninstalled the program on the phone and ran it, now what?
03:04.07tmztListen for connection
03:04.13jerry507_macYou need the console too
03:04.14tmztare you using haretconsole?
03:04.44tmztif you are using linux on pc it is easiest to just un tar it
03:04.52tmztif windows, you have to install python first
03:05.45DracoDanwindows on the desktop...
03:05.54tmztalright, let's use telnet then
03:06.00tmztis it xp or vista?
03:06.14DracoDan:P
03:06.33DracoDani have putty
03:06.46*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
03:07.05tmztputty works
03:07.13tmztconnect to 9999 on your device's ip
03:07.24DracoDando i need the phone on wifi?
03:07.30DracoDanthats a yes...
03:07.47tmztyou just need an ip address for the phone, activesync works, so does bluetooth pan and wifi
03:08.03maejrepI upgraded my kernel the other day, and finally rndis-host works :)
03:08.16tmztmaejrep: what did you upgrade to?
03:08.18maejrepso i can finally connect without using the evdo network
03:08.20maejrep2.6.27 heh
03:08.42tmzthuh, I always use the usb-rndis-host or whatever
03:08.54maejrepevdo is fast and all, but i'm in a concrete building so my signal sucks inside
03:08.58tmztbut it doesn't work with synce-hal
03:08.58maejrepyeah usb-rndis-lite
03:09.19maejrepI haven't been able to get synce working yet, but at least the kernel's built in rndis driver works now
03:09.42tmztI should say it doesn't work with my phone, I think it does work
03:10.54maejrepstill trying to figure out what's wrong with my proc_comm implementation
03:12.35maejrepdo you think I should use an irq handler for proc comm?  It raises the irq, but most of the time the calling function will expect a result back from msm_proc_comm(), which kind of implies that it should wait around for a response
03:13.11tmztmaejrep: did you read what cr2 said earlier?
03:13.33maejrepoh, haven't yet ..  I should read up
03:16.25*** join/#htc-linux divinebovine (n=bite@S01060016b6b53675.vf.shawcable.net)
03:17.38*** join/#htc-linux Heema (n=weechat@217.139.57.79)
03:18.07*** join/#htc-linux Kuma (n=John@g227196088.adsl.alicedsl.de)
03:25.27DracoDanshould haret just have the "working" icon in the middle after hitting "listen for network connections"?
03:25.38tmztyes
03:25.50tmztconnect to port 9999 with putty
03:26.09DracoDanoh, missed the port part :P
03:27.18DracoDanok, all ready
03:27.54DracoDanDetected machine Generic MSM7xxx/MSM7xxx (Plat='PocketPC' OEM='RAPH500')
03:27.54DracoDanCPU is ARM ARM arch 6 stepping 2 running in system mode
03:28.46maejrepDracoDan, what did Device Information say for your processor?
03:29.11maejreptmzt, yeah, I was trying to combine the trout and vogue implementations somewhat
03:29.47DracoDanthe above is what i got on connection
03:29.53DracoDando i need to run a command?
03:30.18maejrepI mean in Start > Settings > Sytem > Device Information > Hardware
03:30.31tmztmaejrep: do we need to get the video mem with VRAM?
03:30.33maejrep(you can do all that with haret still running)
03:30.41maejreptmzt, in what context?
03:30.54DracoDan7500A
03:30.55tmztprint "%08x" VRAM
03:30.55DracoDan...
03:31.07tmztinteresting
03:31.26DracoDanhere it says ram size is 192MB
03:31.34DracoDannormally it says 128MB
03:31.43tmztit says that in haret?
03:31.50DracoDanoh, its split
03:32.00DracoDanno, in Start > Settings > Sytem > Device Information > Hardware
03:32.20tmztok, what do you mean by normally?
03:32.43DracoDannevermind, i see its devided between programs and the OS
03:33.04tmztdepending on what ce calls ram, that's artificial
03:33.06DracoDanOS get 78.71MB, programs get 113.29MB
03:33.35tmztmaejrep: what ram is reported for yours?
03:33.48tmztDracoDan: in putty: print "%08x" VRAM
03:34.08maejrepramsize 288mb, OS=84, Prog = 203
03:34.14DracoDan1686a000
03:36.28DracoDanand that means.....?
03:36.50*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
03:39.41tmztthe address of the video memory, at least for the first part of booting linux
03:40.19DracoDanis there anything else i can do to help?
03:40.39DracoDanim not a programmer...
03:42.32maejreptmzt, mind taking a look at something?  proc_comm command differences, I'm trying to figure out what it means as far as the logic behind each step
03:42.44*** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfd4be.pool.einsundeins.de)
03:42.49maejrephttp://privatepaste.com/c37MKMwzUQ <-- some PCOM command that runs every 10s
03:43.03maejrephttp://privatepaste.com/080PloxouC  <-- something related to inserting the SD card into the slot
03:46.09maejrepa few differences:  1) the SD command requires data, the other doesn't.  2) SD command re-sets the cmd register with |= 0x100 and the other doesn't, 3) Proc sets status register equal to the CMD value in the SD command, but in the other it sets it to 0x100 | CMD
03:46.52maejrepI'm thinking the 0x100 bit in COMMAND tells the processor to expect data in DATA1.  and 0x100 in STATUS register tells the OS that it should expect data in DATA_RESULT1 ?
03:46.53*** join/#htc-linux reformatt (n=chatzill@97-121-31-187.bois.qwest.net)
03:47.53tmztmaejrep: what do you mean by SD command here?
03:48.19maejrepI'm not sure specifically what it's doing, but it happens when I insert the sd card into the slot
03:48.22tmztDracoDan: if you want to try compiling the kernel with that value (but hold on) we can try that but you need linux on the host
03:48.25maejrepthe "other" one just runs every 10s
03:48.56tmztmaejrep: ok, and what do you mean by DATA1, this isn't the smd channel with that name is it?
03:49.25maejrepalso in the SD sequence, its not the OS that resets status to 0 (my thought is its the processor, but not sure), and in the other sequence, the OS sets it to 0
03:49.32tmztmaejrep: is this the same on raph800: VRAM 1686a000 ?
03:49.35maejrepno, DATA1 is the name of the register
03:49.44maejrepsec, i'll check
03:50.28maejrepyes, same
03:50.48tmztok, we should try a raph800 kernel then, is there one available?
03:51.17maejrepyeah latest git should work out of the box, at least with angstrom initrd
03:51.27tmztI mean a zImage
03:51.35maejrepoh a compiled one?
03:52.13maejrepNetRipper has 2 zImages from Jan 4 on his site
03:52.18maejrepnot sure what they're meant for
03:52.42tmztwe can try that, I'm not sure if tekkdrone has
03:52.42maejrepI could compile one and upload it somewhere
03:53.09maejrepmine has the RGB444 fix though
03:53.21maejrepnot sure if raph500 will have the same problem or not
03:53.43tmztif display inits at all I will be happy
03:54.38maejrepthen I'm sure NetRipper's would work
03:55.52tmztmaejrep: could that be a watchdog or keepalive of some sort?
03:56.05maejrepthe every-10s one?
03:56.20tmztyeah
03:56.24maejreppotentially..  its followed immediately by another command with the same procedure, but with 0x8b command
03:56.25tmztnot sure what it is
03:56.40tmztcan you addr2mod ?
03:56.56maejrepI'm more looking for a way to put the apparent procedure into pseudocode logic
03:56.59maejrepsometimes
03:57.13tmztI mean what module is it on ce
03:57.16maejrepI've been able to a few times, but more often than not I get the error about the helper tool
03:57.25tmzttoolhlp?
03:57.30maejrepyeah that
03:57.46maejrep# addr2mod(0xf000fe50)
03:57.46maejrepUnable to create tool help snapshot
03:57.56maejrepi don't know why it does that
03:58.02tmztask about that in #xda-devs, and check what rom you are using
03:58.03maejrepI've seen it work before
03:58.09maejrepstock sprint rom
03:58.21tmzttoolhlp is the debugger helper library
03:58.38maejrepa wince library?
03:58.41tmztyes
03:59.12maejrepah
03:59.15tmztit's specific to the ce version according to cmonex, but she's made one (or found one, not sure which) for mp900c
04:07.02*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
04:14.13tmztDracoDan: sorry
04:14.24tmztDracoDan: can you try a kernel from netripper.nl ?
04:17.59tmzthttp://www.netripper.com/raphael/20090401_zImage_git
04:18.33tmzthttp://www.netripper.com/raphael/initrd-netripper-busybox-usb.cpio.gz
04:19.44*** join/#htc-linux dariball_ (n=dariball@p4FDC4548.dip.t-dialin.net)
04:20.26tmztunfortunately I don't see a default.txt and don't know the mtype
04:20.35tmztmaejrep: what is the mtype?
04:20.42maejrep1910 ?
04:20.51maejrepyeah
04:20.55tmztit's still the same for raph100 and raph800 right?
04:21.02maejrepyes
04:33.21*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
04:58.29tmztDracoDan: are you still here?
05:03.57tmzttekkdrone: ping
05:04.11*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
05:27.07maejrep[    0.440000] msm_proc_comm_wince: waiting for modem; command=0x0000008b data1=0x00000000, data2=0x00000000
05:27.08maejrep[    0.440000] msm_proc_comm_wince: PCOM ready, sending command.
05:27.08maejrep[    0.440000] msm_proc_comm_wince: command and data sent (id=0x2), trying without A2M irq...
05:27.08maejrep[    0.440000] msm_proc_comm_wince: timed out waiting for PCOMCE_CMD_DONE.  retrying with A2M soft-irq
05:27.08maejrep[    0.440000] msm_proc_comm_wince: command status = 0x0000018b
05:27.09maejrep[    0.440000] msm_proc_comm_wince: success. data1=0x00000000, dat
05:27.48maejrepso, looks like the A2M is needed .. but apparently haret never sees it
05:28.23tekkdronetmzt: pong
05:29.06tmzttekkdrone: cool, you have a vzw touch pro?
05:29.20tekkdroneaye :)
05:29.27tmztcan you try these:
05:29.32tmzthttp://www.netripper.com/raphael/20090401_zImage_git
05:29.36tmzthttp://www.netripper.com/raphael/initrd-netripper-busybox-usb.cpio.gz
05:30.09tmztmtype 1910
05:31.05tekkdronesure
05:31.51tekkdroneheh, my browser LOVES that git link
05:31.58maejreplol
05:32.11tekkdroneI wonder what IE would've done >_<
05:32.45tekkdronelemme jump back to nix where I have git installed and all my dev goodies
05:35.47*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
05:39.52maejrephmm, seems like when I try to copy my zImage over scp while booted in linux, it doesn't actually write it
05:40.17maejrepcould just be too large, and running into dma issues (?)
05:40.47tmztdoesn't write it at all or you get errors?
05:41.12maejrepno errors, but when I reboot back to winmo, the file hasn't changed
05:41.23tmztsync
05:41.45maejrep?
05:42.29maejrepam I just unmounting the card too soon, before it finishes flushing its buffers?  usually that would block umount until its finished
05:42.59maejrepbtw, it now boots just as quick as before, and has 2 successful pcom commands :)
05:43.40tmztphysically removing it?
05:43.51maejrepno, I don't remove it
05:43.56maejrepI just umount, then reboot
05:44.03*** join/#htc-linux tekkdrone (n=tekkdron@72.183.115.231)
05:44.07tmztit shouldn't
05:44.19maejrepand when I get back to wm, its the older file still
05:44.31tekkdronetmzt: would you mind repasting those links? :)
05:45.10maejrep-rwxr-xr-x    1 root     root            0 Nov 28 00:12 test-touch
05:45.10maejrep-rwxr-xr-x    1 root     root          101 Nov 28 00:12 test-uname
05:45.30maejrepthose were two files I created when I first got it working, and they're still there, and test-uname still has the correct data in it
05:45.37tmzthttp://www.netripper.com/raphael/20090401_zImage_git
05:45.39maejrepmy guess its related to the size of the zImage file
05:45.41tmzthttp://www.netripper.com/raphael/initrd-netripper-busybox-usb.cpio.gz
05:45.45tmztmtype 1910
05:46.09maejrepso, anyone have any ideas how to reverse engineer the pcom commands? :)
05:46.25tekkdronenot a clue here, I haven't even glanced at the newer archs
05:49.14*** join/#htc-linux mickey (n=mickey@e180159077.adsl.alicedsl.de)
05:50.03tmztmickey: are you usually in #android?
05:50.17maejrephmm
05:50.28maejrep<PROTECTED>
05:50.28maejrep<PROTECTED>
05:50.28maejrep<PROTECTED>
05:50.28maejrep<PROTECTED>
05:50.28maejrep<PROTECTED>
05:50.29maejrep<PROTECTED>
05:50.31maejrep<PROTECTED>
05:50.33maejrep<PROTECTED>
05:50.37maejrepthat looks like the same on my device
05:50.53*** join/#htc-linux goxboxlive (n=goxboxli@mail2.hjellnesconsult.no)
05:51.35tmztmaejrep: where is that from?
05:51.40maejrepvogue-smd.c
05:51.55maejrepi'm positive that is how my smd AT buffer works :x
05:52.04maejreplooking at all the same memory offsets, they match
05:52.23tmztdo you see the head/tails changing at those addresses?
05:52.32maejrepi can trace it
05:52.58maejrepbut I just checked an old dump, and I see data where I expect to see it, and the offset matches
05:53.10maejrep(matches the split in the buffer)
05:53.22maejrepi'll have to reboot to wm to trace for sure
05:53.49tmztmaejrep: have you tried the vogue smd?
05:54.16maejrepno, think I should just replace it, or try to port the hard-coded 0 and 1 channels?
05:54.41maejrephis entire structure is quite different
05:56.31tmztnot sure, what do you mean by hard-coded 0 and 1 channels?
05:56.38tmztjust the registration of those?
05:56.53tmztsmd_alloc ?
05:56.56maejrepin smd_open
05:57.06maejrephttp://git.linuxtogo.org/?p=groups/mobile-linux/kernel.git;a=blob;f=arch/arm/mach-msm/vogue-smd.c;h=61482d06cef55edbb5dc21423b09bd5721c9f84d;hb=refs/heads/htc-vogue#l594
05:57.33tmztso the addresses are hard-coded for those channels?
05:58.00*** join/#htc-linux pleemans (n=toi@d54C2AAB7.access.telenet.be)
05:59.48maejrepyes
05:59.48tekkdronetmzt what's the kernel cmdline for this setup?
06:00.12tmztit should work without it
06:00.34tekkdronehmmm
06:00.45*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
06:00.57*** join/#htc-linux zule (i=foglem@shadowmite.com)
06:01.00maejrepwhat doesn't work about it?
06:01.14tekkdronethe kernel starts but hangs at console handoff to tty0, figured cmdline needed a tweak
06:01.46maejrepyou've booted a kernel on this phone before?
06:01.52tekkdronenegative
06:01.56maejrepugh
06:02.06maejrepi hope it's not another panel difference :x
06:02.15tekkdroneha.
06:02.24tekkdronewe
06:02.33tmztcan we get usb to work, that should at least tell us if it has hsusb
06:03.03tekkdroneif it stops output at the console handoff again I"ll wait 60 seconds or so and hook it up
06:03.05maejrepit  won't be initialized until the initrd loads
06:03.38tekkdronewhat is this initrd setup expecting for root?
06:03.41maejrepi think you might have to have it plugged in at boot time for it to work (that's what I recall NetRipper saying .. not sure if that's still the case)
06:03.42tekkdronekeepinitrd?
06:03.55tekkdronemaejrep: lol just like my progress on apache
06:05.49tekkdroneok, attempting with: set mtype 1910
06:05.50tekkdroneset kernel "zImage"
06:05.50tekkdroneset initrd "initrd"
06:05.50tekkdroneboot
06:06.00tmztok
06:06.07tmztwhat did you do last time?
06:06.16tekkdronesame + cmdline=tty0
06:06.25tekkdroneerr, console=tty0
06:06.38tmztif you have usb storage or advanced networking enabled, disable it
06:06.53tmztconsole=tty0
06:08.46*** join/#htc-linux jerry507_mac (n=jerry507@12-226-245-130.client.mchsi.com)
06:12.28maejrepyeah, the usb storage thing is what got me :/
06:12.42maejrepif it asks you what mode to use when you plug in USB, make sure it's in activesync mode
06:12.53maejrepadvanced networking can be enabled or disabled, doesn't matter
06:15.54*** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz)
06:16.04tekkdroneusb storage and rndis off, blank cmdline, last line seen: console handover: boot [htc_fb-1] -> real [tty0]
06:16.24tmztok, what does lsusb -v -v show on pc?
06:16.27tekkdronewaiting a few in case we're just not seeing output, booted connected to usb
06:16.39tmztyou should see two CDC ethernet interfaces
06:16.50tekkdronelsusb -v -v
06:16.50tmztdon't reboot it
06:16.52tekkdrone?
06:16.57tmztyes
06:17.00tekkdronesec
06:18.18tekkdroneBus 002 Device 012: ID 0bb4:0a19 High Tech Computer Corp.  with two bulk transfer EP's
06:18.42tekkdronehowever, after bInterval 0, I see: can't get device qualifier: Connection timed out
06:18.42tekkdronecan't get debug descriptor: Connection timed out
06:18.43tekkdronecannot read device status, Connection timed out (110)
06:22.02tmztyeah, that doesn't sound good
06:22.23tekkdronesounds like the 500 has a different panel then no?
06:22.30tmztI mean the usb
06:22.46tekkdronemaejrep was worried that might the case on the video
06:23.31tmztit might be, I mean the usb should have worked if it's similar to msm7201a/7501a
06:23.59tmzt"connection timed out"
06:24.11tekkdronethat was after pulling most of the information
06:25.08tekkdronethe fact the device provided endpoints without the kernel puking all over my screen is promising
06:25.12tekkdronethey can't be all that different
06:25.31tekkdronenot that I would've seen the kernel printk's....
06:26.52tmztmaejrep: can we disable the console switch and see what's going on?
06:27.04maejrepyeah
06:28.30maejrephmm
06:29.32maejrepcan't find it :|
06:29.41*** join/#htc-linux pichurri (n=pichurri@194.230.146.121)
06:29.51maejrepCONFIG_HTC_FB_CONSOLE=y
06:29.52maejrepCONFIG_HTC_FB_CONSOLE_DELAY=y
06:29.52maejrepCONFIG_HTC_FB_CONSOLE_BOOT=y
06:30.00maejrepunset FB_CONSOLE_BOOT
06:30.18tmztand disable FB_CONSOLE?
06:30.20maejrepand iirc, console=ttyMSM0
06:30.24tekkdronebtw here's the device lsusb output http://pastebin.com/d6e00b812
06:30.25maejrepno, keep fb_console
06:30.48maejrepHTC_FB_CONSOLE is druidu's console, what he sees initially
06:31.01maejrepFB_CONSOLE_BOOT=y means "only use it during boot"
06:31.06tmztsorry, I mean FRAMEBUFFER_CONSOLE
06:31.13tmztCONFIG_FRAMEBUFFER_CONSOLE
06:31.13maejrepso disabling that makes it continue to use druidu's console
06:31.19maejrephmm, possibly
06:31.21maejrepsec
06:32.27tmztlsusb -v -v -d 0bb4:0a19
06:34.44tekkdroneand man... haret loads sooooooo much faster on these that my apache
06:35.51maejrep<NetRipper> under system type -> boot console for htc phones -> disable the 'use console only during boot'
06:36.10*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
06:37.14tekkdronetmzt: same output on the device-targeted verbose lsusb
06:37.25tmztso you get the same errors?
06:38.17tekkdroneyes to stderr, they aren't captured by lsusb -v -v -d 0bb4:0a19 > output
06:39.10tmztthis is a different vend/prod than in wm?
06:39.28tekkdronegood question, sce
06:39.31tekkdrone*sec
06:39.39tmztthat initrd should have brought up the usb0 interface
06:40.05tekkdronewhich mode in WM? cdc?
06:40.40tmztit should be serial if advanced networking is disabled
06:40.55tmzttry booting with advanced networking enabled
06:41.45tekkdronesame ID with WM attached
06:42.33tmztah
06:43.15tmztdoes your device support usb 2.0?
06:43.16tekkdronelsusb output is almost identical to wm without adv. networking
06:43.53tmztI think it will show in dmesg as either uhci/ohci or ehci
06:44.19tekkdroneehci
06:44.24tekkdronehigh speed
06:44.45tekkdroneInteresting: [ 2498.752771] usb 2-2: usbfs: USBDEVFS_CONTROL failed cmd lsusb rqt 128 rq 6 len 255 ret -110
06:44.59tmztyeah
06:45.56tekkdronebut yes, the linux lsusb and windows non-adv-net lsusb output are identical, except polling the device on netripper's kernel cuts off the Device Qualifier section
06:47.02tmztcat /sys/bus/usb/devices  bNumInterfaces for the usb device listed in lsusb -v -v
06:48.25tekkdroneunder which configuration
06:48.35tekkdronewm with net, without net, or netripper's image
06:48.48tmztin netripper's image
06:50.58tekkdrone1
06:51.29tekkdronesorry, bNumInterfaces = 1
06:51.30tmztand that's the same as windows?
06:51.37tmztwm
06:51.42tekkdronetell you in 30 seconds
06:51.44tmztnon-rndis
06:51.55tekkdrone:)
06:52.24tekkdroneha, I'd be @ 55-60% battery on apache already
06:52.32tekkdronereboots = death
06:54.09tekkdronesame in WM, but the kernel picks WM up right away with : [ 4677.445169] ipaq 2-2:1.0: PocketPC PDA converter detected
06:54.09tekkdrone[ 4677.445868] usb 2-2: PocketPC PDA converter now attached to ttyUSB0
06:54.49tmztmaejrep: what addresses does usb use on raph100/800, can we trace those with haret?
06:55.04tmztjust the registers, not the ulpi stuff right now
06:55.30maejrepi know its irq 47
06:55.48maejrepnot sure about the others
06:55.57tekkdronehmmm hang on a sec
06:55.59maejrepthe memory map should be fixed on the wiki now
06:56.09maejrep(had to split it into 3 pages for it to not white-screen)
06:57.07*** join/#htc-linux kiozen (n=oeichler@p5492A377.dip0.t-ipconnect.de)
07:00.04tekkdronetmzt: this looks like what I would see on apache, if I didn't replug USB the udc reset wouldn't occur and linux would just see the old stale usb settings from WM
07:00.39tekkdroneI did a tail -f on /var/log/messages while the netripper kernel came up, no usb activity whatsoever
07:01.33tekkdroneso I retract my earlier statement about the device providing endpoints
07:01.42tekkdronelol, we're probably just seeing stale ones from WM :(
07:01.59tmztright
07:02.21tmztand it can't read from the control endpoint when run lsusb (of access usbdevfs)
07:02.24tmztor
07:03.06tekkdronebummer
07:05.21tekkdronelet it be known that my machine state at haret runtime is not stock and I've got hardSPL on here
07:05.39*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
07:05.43tekkdronevzw gimps their roms too much...
07:07.10*** join/#htc-linux ecze (n=ecze@eczema.ecze.com)
07:09.04*** join/#htc-linux ecze (n=ecze@eczema.ecze.com)
07:21.16*** join/#htc-linux rl2000 (n=rl2000@0x573b8b62.hhnqu1.dynamic.dsl.tele.dk)
07:22.25*** join/#htc-linux tekkdrone_ (n=tekkdron@72.183.115.231)
07:23.06tekkdronewas the last thing you saw me say about vzq gimping their roms?
07:23.47tmztyes
07:23.51tekkdrone>_<
07:24.11tekkdronemaejrep: any quick panel traces for verification?
07:24.43tekkdronetmzt: usb 2-2: new high speed USB device using ehci_hcd and address 30 is seen when replugging the device running netrip's kernel
07:25.06tekkdronethen it retries 3 times and stops
07:25.30maejrepi get:
07:25.30tmztand lsusb doesn't work?
07:25.31maejrepusb 1-5: new high speed USB device using ehci_hcd and address 59
07:25.31maejrepusb 1-5: configuration #1 chosen from 1 choice
07:25.31maejrepusb0: register 'cdc_subset' at usb-0000:00:13.5-5, Linux Device, 62:e3:f5:82:d0:fb
07:25.31maejrepusb1: register 'cdc_subset' at usb-0000:00:13.5-5, Linux Device, 62:e3:f5:82:d0:fb
07:25.40tekkdroneright, the device is not present in lsusb
07:25.56tmztyou get error -81 ?
07:26.05*** join/#htc-linux rmoravcik (n=rmoravci@gtsgw.ttc.cz)
07:26.33tekkdroneno errors in dmesg or on lsusb
07:26.44tmztwhat do you mean tries 3 times?
07:27.03tekkdroneusb 2-2: new high speed USB device using ehci_hcd and address 30
07:27.05tekkdroneusb 2-2: new high speed USB device using ehci_hcd and address 31
07:27.12tekkdroneetc up to 33
07:27.25tmztbut it doesn't timeout or anything else?
07:27.29tekkdronenope
07:27.36tekkdroneweird huh?
07:27.42tmztyeah
07:27.49tmztdo you have any other usb 2.0 devices connected?
07:28.05tmztincluding hubs
07:28.08tekkdronemmmmmm
07:28.28tekkdronemy kbd/mouse (single plug) and a usb key
07:29.24tekkdroneare the only two devices besides the raph, the usb key is 2.0 compliant if i remember correctly
07:29.44tmztcan you unplug that? the kbd/mouse shouldn't matter
07:29.47tekkdroneyep, and now sitting on my desk
07:29.53*** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfe2db.pool.einsundeins.de)
07:30.05tekkdroneit was usb 2-5, address 3
07:31.10tmztalright, modprobe -r ehci-hcd
07:31.20tmzttry attaching the raph in linux
07:31.43tmztcheck dmesg and see how many devices are registered
07:33.19*** join/#htc-linux BabelO (n=fcr@unaffiliated/babelo)
07:34.00tekkdronehmmmm there's still a 2.0 device on this bus
07:34.07tekkdroneinternal wlan AP
07:34.38tekkdroneWM connected at full-speed though rather than hi-speed so I think we achieved the desired effect
07:35.20tmztok, still not in lsusb?
07:35.35tekkdronejust now booted ripper's kernel, sec
07:35.44*** join/#htc-linux Nullbyte (i=nullbyte@85.17.58.168)
07:36.07tekkdronesame time-outs on lsusb
07:36.18tekkdroneI do have an ohci internal on this box, going to try that
07:36.32*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
07:36.56NullbyteHi guys :)
07:38.16tekkdronetmzt: incrementing addresses again under usb1
07:38.20tekkdroneNullbyte: hello
07:38.43tmztusb1?
07:39.28NullbyteHey :) DO you guys have any idea if it it's possible to use something like HaRET on a qtek s200?
07:39.39Nullbyteiirc its similar to the htc prophet
07:39.52tekkdroneusb 1-2: new full speed USB device using ohci_hcd and address 10, 11, 12 etc
07:39.54tmztwhat is that? can you find it on wiki.xda-developers.com ?
07:40.03tmztoh prophet?
07:40.04Nullbyteor uh, is the same thing
07:40.12Nullbyteapparently
07:40.25tmztit's omap850-based?
07:40.46Nullbyteyep
07:40.57tmzttry the linwizard kernel
07:41.05Nullbytethanks, ill look at it :)
07:41.24tmztif the display doesn't work, that can be fixed
07:42.02NullbyteI'll join their irc channel then, thanks
07:47.33tmztthey are not really there right now Nullbyte
07:50.39tekkdronetmzt: different linux distro/hardware netted some different results
07:50.41*** join/#htc-linux pichurri (n=pichurri@users3.ilo.org)
07:51.08tekkdroneusb 3-4: new high speed USB device using ehci_hcd and address 5, Jan  8 03:50:01 linux-x59u kernel: usb 3-4: device descriptor read/64, error -110,  Jan  8 03:50:01 linux-x59u kernel: usb 3-4: device descriptor read/64, error -110
07:51.08tekkdroneJan  8 03:50:16 linux-x59u kernel: usb 3-4: new high speed USB device using ehci_hcd and address 6
07:51.44*** part/#htc-linux paulproteus (n=paulprot@2002:db69:2513:0:0:0:0:1)
07:52.05tmztread64: error -110, that's what I meant earlier
07:52.23tekkdroneyep, i'm seeing the error on this other machine
07:53.07tmztand the one that works, lsusb works now?
07:53.24*** join/#htc-linux paulproteus (n=paulprot@2002:db69:2513:0:0:0:0:1)
07:54.07tekkdroneno, still a ghost entry
07:54.21tekkdroneit does that same address incrementing shiz on the other box
07:55.17tekkdronea different one now, device not accepting address errors
07:55.43tekkdrone(combined with the incrementing addresses)
08:04.00*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
08:06.23*** join/#htc-linux Zinbolic (n=Zy@84.238.80.215)
08:27.31*** join/#htc-linux BabelO_ (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net)
08:30.25br1ckk guys, who broke the xda forum :P
08:33.00maejrepok, smd hacked with vogue
08:33.03maejrepjust need to test it
08:34.02*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
08:40.03*** join/#htc-linux Jerry507 (n=jerry507@12-226-245-130.client.mchsi.com)
08:53.08*** join/#htc-linux Jerry5071 (n=jerry507@12-226-245-130.client.mchsi.com)
08:56.20*** join/#htc-linux timebomb (n=tb@e176126245.adsl.alicedsl.de)
09:02.19NetRippermaejrep, http://wiki.xda-developers.com/index.php?pagename=RaphaelDEX
09:02.45maejrepnice
09:03.03NetRippertekkdrone, you have to boot with activesync mode on and usb cable connected
09:03.12maejrepis that the pcom data?
09:03.15NetRippermaejrep, yes
09:03.27NetRippercr2 reversed that
09:03.31maejrepsweet
09:03.55*** join/#htc-linux iNFEKT (n=nFKT@76.20.185.254)
09:04.20NetRippermaejrep, did you try the vogue implementation of proc_comm?
09:04.41maejrepits very similar
09:04.43NetRipperor is your proc_comm working now? (and is it with or without irq?)
09:04.57maejrepmine is working with the A2M irq (timed out without it)
09:05.07maejrepbut not hooking M2A irq
09:05.29NetRipperwasn't the A2M the problematic one?
09:05.39maejrepwasn't problematic, no
09:05.42NetRipperok
09:05.43maejrepjust wasn't showing up in haret
09:06.03NetRipperso what's different fromt he original proc_comm.c? that you don't use M2A?
09:06.06maejrepand i *think* it plays safely with the trout proc_comm too
09:06.18maejrepthe original proc_comm doesn't use m2a either ;x
09:06.23maejrepgo figure
09:06.34NetRipperok lol so then what did you do to get it working ;p
09:06.52maejreprewrote it? :p
09:07.11maejrepit basically works like vogue's does, in the sense that we use PC_COMMAND and PC_NUMBER, etc
09:07.17NetRipperok
09:07.19maejrepwhere trout's uses APP_* and MDM_* registers
09:07.25NetRipperah, ok
09:07.54maejrepbut I think this works slightly more effectively with commands that send data to proc comm
09:08.26NetRipperalright good job
09:08.28maejrepi actually don't know if what I did was correct, but it works with the two commands I tried
09:08.47NetRipperyou could try vibra motor to see if arm9 really processes it :)
09:08.49maejrepI'll try it with vib, that's an easy one to tell if its working or not :)
09:08.51maejreplol
09:08.58NetRipperlol
09:09.13NetRipperright.
09:09.47maejrepso is that just 0x17, delay, then 0x18?  don't need to send data to it (like 1 for on)?
09:10.17*** join/#htc-linux Jerry507 (n=jerry507@12-226-245-130.client.mchsi.com)
09:13.19maejrepso whether this works or not i need to go to sleep :x
09:13.56maejrepthe smd hack didn't work btw
09:14.10maejrepbut the vib pcom did :D
09:14.24maejrep<PROTECTED>
09:14.24maejrep<PROTECTED>
09:14.24maejrep<PROTECTED>
09:22.57NetRipperi dont think you need data
09:23.09NetRipperand, nice
09:23.10NetRipper:)
09:23.47maejrephttp://privatepaste.com/dc16uyycGg
09:24.08NetRippernice
09:24.17maejrepits mostly a copy of proc_comm.*, except proc_comm.h's constants are removed, since we have no use for them
09:24.40NetRipperthough we can just call it msm_proc_comm() and use the Makefile to use proc_comm_wince.c instead of proc_comm.c
09:24.49NetRipperok
09:25.26maejrepyeah, there was something in vreg that needed something in proc_comm, and I didn't spend any time looking into why
09:25.33maejrepso that's why I left proc_comm.o in the make file
09:25.44maejrepand made proc_comm_wince per cr2's comment
09:26.08NetRipperok
09:26.36NetRipperyes but to prevent a lot of #ifdefs for HTCRAPHAEL all over the other files, the Makefile is an easier way it hink ;)
09:26.48maejrepagreed
09:27.27maejrepjust wait till I start complaining that raph800 needs a bunch of ifdefs ;p  (*cof*smd*cof*)
09:27.32NetRipperdo you have git access btw?
09:27.36maejrepno
09:27.43NetRipperok
09:29.38maejrepok, time for sleep
09:29.46NetRippergood night :)
09:30.28*** join/#htc-linux kilian_ (n=kilian@92.66.94.81)
09:32.15*** join/#htc-linux imfloflo (n=imfloflo@cap31-6-88-180-73-121.fbx.proxad.net)
09:34.02lupine_85ho de hum, I need to work out .bb files now
09:38.30imflofloxda still down :s
09:39.22*** join/#htc-linux kimhoon (n=kimhoon@s559116c1.adsl.wanadoo.nl)
09:41.56NetRippermaejrep, at what point do you call the init? in board?
09:59.30NetRippermaejrep, works for me on raph100 as well
09:59.52NetRipperany diam100/diam800 users about?
10:01.32*** join/#htc-linux dcordes (n=dcordes_@unaffiliated/dcordes)
10:01.55kimhoonNetripper--> on what??
10:02.18NetRipperkimhoon, ah, you got a diamond, right?
10:02.26*** join/#htc-linux Mullins (n=bw@89.204.250.156)
10:02.36kimhoonnetripper --> jep
10:02.42NetRipperkimhoon, can you try a kernel for me?
10:02.49NetRipperhttp://www.netripper.com/raphael/zImage
10:02.59NetRipperuse mtype 1910 (raphael mtype)
10:03.29NetRipperthe phone should vibrate for 5 seconds during linux boot
10:03.41dcordesNetRipper did you test the diamond machine you put in the trr yet?
10:03.49NetRippertrr?
10:04.09NetRipperdcordes, the one in git? yes, the diamond board works
10:04.41NetRipperbut for this kernel i havent changed the diamond board, so kimhoon needs to test with raphael mtype ;)
10:05.57kimhoonand its vibrate :) funny :)
10:07.36NetRippernice
10:07.41NetRipperdid you have diam100 or 800?
10:07.59kimhoondiam100 i'm from holland :)
10:08.09NetRipperah right, forgot
10:08.11NetRipper;)
10:08.28NetRippermaejrep, fyi, proc comm working on raph100 and diam100 as well
10:11.33j0b0rotating android to landscape when raph keyboard is open works :) (also on other input devices than event0)
10:11.52NetRippersweet
10:11.59NetRippernice :)
10:12.05dream_killhi
10:12.08NetRipperwe're making good progress now ;)
10:12.08dream_killnew kernel ??
10:12.16j0b0the menu tab still slides from navipad to speaker, which is now from right to left
10:12.40j0b0but the signal/battery/clock sit on what used to be the right and is now the top
10:13.11NetRipperim having trouble visualizing ;)
10:13.26j0b0this will likely turn diamonds on their sides permanently
10:13.35NetRipperyes
10:13.45NetRipperyou need the machine_is_htcraphael() around it
10:13.55*** join/#htc-linux Abracadabr4 (n=aaabraca@62-244-191-249.cust.exponential-e.net)
10:14.04NetRipperso that it only works on raphael mtype
10:16.41j0b0this one does not have machine_is_raphael() around it --> http://tibook.jb.b4m.com/j0b0/zImage-rotate-jobo
10:17.01j0b0but it should help you visualize on your raphael
10:19.04j0b0sorry .. that was 403
10:19.13*** join/#htc-linux milfadoodle (n=milfadoo@78-69-144-82-no19.tbcn.telia.com)
10:19.25dream_killmmm
10:19.29dream_killusewd ur kernel
10:19.40dream_killat bootup it connects the buzzer :D
10:19.46dream_kill(netripper)
10:20.19j0b0really? :D do we have the vibrator working?
10:20.39j0b0does that mean we can also control LEDs and backlight?
10:21.19dream_killbut what is strange
10:21.28dream_killis the radio i have in device now is VERY old
10:21.32dream_kill(MFG one)
10:21.43dream_killand the mtty detected now as a GENERIC MSM ..:P
10:21.46dream_killnot raphael :D
10:22.42imflofloxda up :)
10:23.03j0b0be gentle with it ;)
10:23.51NetRipperj0b0, no, maejrep implemented proc_comm, and we just tested if it worked by enabling vibra motor via proc_comm
10:24.00NetRipperso proc_comm is operational now
10:24.41NetRipperso unless led/backlight work via proc comm? :)
10:24.51NetRipperbacklight might actually, when i look at proc_comm commands
10:25.02NetRipperbut just on/off
10:25.25*** join/#htc-linux StarLiteMobile (n=pocketir@84.241.204.131)
10:26.41kimhoonbut the light sensor works if you know that?? on diam
10:27.14dream_killnetripper, with ur kernel, my device no linux anymore :D
10:27.22dream_killwhat radio do you have in ur device :P ?
10:27.25NetRipperlight sensor does not work yet afaik
10:27.44NetRipperdream_kill what do you mean no linux?
10:27.55dream_killbefore it booted up
10:28.00dream_killnow it stuck at the sd card
10:28.05NetRipperyou just said buzzer worked? :)
10:28.10dream_killyes
10:28.14dream_killwhen booted up
10:28.19dream_killthe buzzer started :D
10:28.49NetRipperah ok, that sd card doesn't work might be because you're using raphael mtype instead of diamond mtype
10:28.59NetRipperthis was just an experimental kernel
10:29.08dream_killit worked before with the config i had
10:29.17dream_killnow i have a different radio in
10:29.30NetRippergo back to a kernel that did work for you for sur
10:29.38NetRipperdont use this kernel
10:30.22dream_kill2 sec
10:30.26*** join/#htc-linux metter (n=metter@242.68.202.62.cust.bluewin.ch)
10:30.33dream_killwriting the original kernel which worked before
10:30.58dream_killbut is starnge as Hatter doesn't detect the phone now as raphael anymore :D
10:31.11NetRipperanyway im not using an mfg radio if that's what you mean
10:31.12NetRipper:)
10:31.15NetRipperoh
10:31.20NetRipperharet
10:31.25dream_killyes
10:31.32NetRipperdoes it detect MSM7xxxA though? that's the important part
10:31.32dream_killI'M using MFG radio :D
10:31.50dream_killbefore used a normal radio and detected as raphael :D
10:31.55dream_killyesterday i flashed a MFG image :D
10:32.05dream_killto check something in bootloader
10:32.13NetRipperaha
10:32.26NetRippercould be incompatible
10:33.10dream_killFound machine Generic MSM7xxxA
10:34.20dariball_thats the same for my branded diam (germany o2) ... works fine
10:34.38lupine_85yay, moving forward with the angstrom compile
10:35.02NetRipperyea some roms use a different identification string
10:35.16NetRipperwhich makes it not detect it as raphael, but as msm7xxxa, it's not a problem
10:35.24NetRipperhad it with elite rom as well
10:35.32dream_killit booted up
10:35.36dream_killfinaly with ur kernel
10:35.52dream_killstarted now init.android
10:35.54dream_killlet's see it :D
10:37.19lupine_85is the defconfig_htcraphael available on the gitweb?
10:37.51dream_killbut keyboard now
10:38.00dream_killnot working still?
10:39.06dream_killit detects the slide, but not rotate screen
10:39.17dream_killit behave like backspace
10:41.41*** join/#htc-linux dcordes_ (n=dcordes_@unaffiliated/dcordes)
10:41.47StarLiteMobileso, imagine I want to get a new phone to replace my old tytn, with the intention to run android on it. Which one is the most likely to ever run it?
10:42.06StarLiteMobilexept the G1 offc :P
10:42.32dcordes_StarLiteMobile any msm7xxxA based
10:43.06StarLiteMobilewhich ones are that? touch pro? touch hd?
10:43.19dcordes_StarLiteMobile yep
10:43.25dcordes_xperia x1
10:43.32StarLiteMobilexperia?
10:43.33dcordes_diamond
10:43.33StarLiteMobilecool
10:43.49dcordes_yea. xperia is a nice device
10:44.13StarLiteMobilereally in doubt atm. I really want to lose winmobile, but the G1 isn't really for me
10:44.26dcordes_you need some patience
10:44.51dcordes_if youw ant to use one of them as a primary cell phone
10:45.02StarLiteMobilein doubt tween the X1, touch hd and pro. dont want to lose the keyboard, but that 3.8 inch screen of the hd is soo nice
10:45.19lupine_85touch pro keyboard is amazing
10:45.23lupine_85if we ever get it working
10:45.36j0b0dream_kill the screen rotation is only in the kernel i just linked to in here. i guess youre booting netrippers vibrator kernel?
10:46.02StarLiteMobilei like the hermes/tytn kb.
10:46.20StarLiteMobilei tried an iphone and it kinda sucked kb wise
10:46.33dcordes_touch pro keyboard is likea  suped up tytn keyboard
10:47.40StarLiteMobileif only my hermes would be faster.. i would wait for the G2
10:48.00StarLiteMobilebut its really starting to annoy me now
10:49.13NetRipperj0b0, you should consider getting in sync with git again ;)
10:49.29dcordes_I think all of the new htcs are very fast
10:49.42NetRipperwinmo is sluggish though :s
10:50.41StarLiteMobilei heard the iphone cant multitask.. why would people *ever* want one? :S
10:50.54lupine_85the hardware? or the OS?
10:50.55StarLiteMobilei always have mutiple things open..
10:51.04StarLiteMobileif only I had more ram..
10:51.08StarLiteMobilethe os prolly
10:51.08dcordes_what do you mean, it can't multitask?
10:51.21lupine_85if it has an MMU, you can get it to multitask, I'm sure
10:51.40StarLiteMobilewell, you cant have mutiple programs open at once
10:51.48StarLiteMobileeg: google mas & browser
10:51.59StarLiteMobileit will kill one if you open the other...
10:52.18dcordes_ah
10:52.41dcordes_that isa  question of user interface
10:53.20dcordes_nowadays you can boot linux on the iphone devices
10:53.29dcordes_I think it's not so popular
10:53.57dcordes_but when there is breaking news: googel android booted everboy will want one
10:53.58StarLiteMobilewhat about android?
10:54.03StarLiteMobileon the iphone
10:54.12StarLiteMobileyeh
10:54.26StarLiteMobileor well. id still get a touch hd
10:54.33StarLiteMobilefor the resolution
10:54.45StarLiteMobile800*640 baby :D
10:55.15StarLiteMobilevs the iphones crappy 320*480 or so
10:55.45NetRipper480x800, isnt it?
10:55.52lupine_85touch pro is 480x800
10:55.56lupine_85IIRC
10:56.00StarLiteMobileAh
10:56.01NetRippertouch pro is 480x640
10:56.05lupine_85is it?
10:56.07NetRipperyes
10:56.07StarLiteMobilealmost correct then
10:56.11dcordes_raphael has vga.
10:56.12lupine_85aww
10:56.13lupine_85:D
10:56.38NetRippersorry i just destroyed your world lupine
10:56.39NetRipper;)
10:56.44StarLiteMobilex1 is wvga as well
10:56.54NetRipperyes
10:57.45NetRipperand iphone multitask is by choice of apple in the os
10:57.55NetRipperit is to prevent the device from getting sluggish when too many apps are open
10:58.07NetRipperand to prevent needing a task manager of some kind
11:00.05lupine_85fail on their part
11:00.13lupine_85NetRipper: I'll cope :)
11:00.21lupine_85640x480 is easier to cope with anyway :)
11:00.47lupine_85it just looks wider because of the windows task bars, I suspect
11:00.58lupine_85gives it a vaguely widescreen aspect
11:00.58*** join/#htc-linux trn_ (n=trn@128-242-dsl.kielnet.net)
11:02.57dcordes_wide screen is in
11:03.13lupine_85makes it look like it should be widescreen, even though it isn't
11:03.19dcordes_hip
11:03.35dcordes_the xperia has keyboard AND widescreen
11:03.46lupine_85the keyboard hardware is inferios
11:03.49lupine_85inferior*
11:03.49NetRipperi played with a touch hd and the larger res is pretty sweet
11:04.04lupine_85touch pro keyboard is more pleasant to use
11:04.09dcordes_but the x1 keyb is opticaly better than practically
11:04.21NetRipperat least there you can get multiple appointments on the today screen instead of having to scroll to see the second one ;)
11:04.24lupine_85(xperia feels a bit cheap all round, really)
11:04.31lupine_85opticaly better?
11:04.44StarLiteMobilei like the looks
11:05.05StarLiteMobilehardware wise the diff tween the touchpro and x1 are small
11:05.13lupine_85StarLiteMobile: it looks prettier, true... but for me that fell down the moment I touched it
11:05.16dcordes_lupine_85 I never used raph keyboard but I prefer kaiser over venus (x1) keyboard
11:05.30lupine_85the touch pro keyboard is so much easier to type with
11:05.46dcordes_does venus have magnetrometer?
11:05.48StarLiteMobileis the x1 kb on par with the tytn kb tho?
11:06.07*** part/#htc-linux trn_ (n=trn@128-242-dsl.kielnet.net)
11:06.23dcordes_StarLiteMobile no. the x1 keyboard rows have offsets likea  normal keyboard
11:06.25StarLiteMobilei think both x1 and tpro dont have accel meter do the?
11:06.35dcordes_and there is some space between the kys
11:06.44StarLiteMobilehtc never use those on hardware kb phones
11:07.01StarLiteMobilemzz
11:07.17dcordes_StarLiteMobile raphael has magnetrometer
11:07.18StarLiteMobilereally have to try it out in the shop then
11:07.46lupine_85StarLiteMobile: just make sure you touch them both :)
11:07.55StarLiteMobileyeh
11:07.56lupine_85(touch pro was also 1/2 the price for me :D)
11:08.01lupine_85£50 rather than £100
11:08.13StarLiteMobilei wish the touch hd came with a kb.. or android :P
11:08.14dcordes_StarLiteMobile I really liked x1 look and feel. but the kyboard sure neds some practice
11:08.59dcordes_with a 2 yea contract, vodafone gemrnay wants  350 eur for raphael
11:09.35dcordes_StarLiteMobile x1 resolution = hd res.
11:09.36StarLiteMobileare there any decent android phones on the horizon btw?
11:09.55dcordes_StarLiteMobile you mean with android shipped?
11:10.03StarLiteMobileyeh
11:10.24StarLiteMobilesamsung had plans, didn't they?
11:11.37dcordes_no clue I'm not very convinced by android
11:11.47*** join/#htc-linux myxor (n=myxor@pdbn-4d089087.pool.mediaWays.net)
11:12.08StarLiteMobileit heaps better then winmo imho
11:12.51dcordes_it runs on linux
11:13.24dcordes_imho the downside is the licensing situation
11:13.26dcordes_no gpl
11:13.55StarLiteMobilestill better as closed source tho
11:14.07dcordes_and the subsystems interfacing the hardware hve no sourcesa t all
11:14.28dcordes_ie these prts are closed source
11:14.43StarLiteMobiletrue
11:14.55StarLiteMobilemakes devving harder then needed
11:16.18dcordes_I think a good indicator is that the actual piece of software whch makes the phone a phone is closed source
11:16.43dcordes_StarLiteMobile always depends on what you consider deving
11:17.12StarLiteMobileprolly phone manufacturers to blame there tho
11:17.27dcordes_if you are happy making some java apps that run in a box it's all good. but android doesn't allow for fundamental changes
11:17.42dcordes_in the system which makes it very static
11:18.45dcordes_StarLiteMobile no, rather operators I think
11:19.03dcordes_(to blame for the closed modem server)
11:19.29StarLiteMobileas well
11:20.26StarLiteMobileoo lunch time
11:20.30StarLiteMobilebrb :P
11:39.21*** join/#htc-linux ChanServ (ChanServ@services.)
11:39.21*** join/#htc-linux myxor (n=myxor@pdbn-4d089087.pool.mediaWays.net) [NETSPLIT VICTIM]
11:39.21*** join/#htc-linux dcordes_ (n=dcordes_@unaffiliated/dcordes) [NETSPLIT VICTIM]
11:39.21*** join/#htc-linux metter (n=metter@242.68.202.62.cust.bluewin.ch) [NETSPLIT VICTIM]
11:39.22*** join/#htc-linux StarLiteMobile (n=pocketir@84.241.204.131) [NETSPLIT VICTIM]
11:39.22*** join/#htc-linux milfadoodle (n=milfadoo@78-69-144-82-no19.tbcn.telia.com) [NETSPLIT VICTIM]
11:39.22*** join/#htc-linux Mullins (n=bw@89.204.250.156) [NETSPLIT VICTIM]
11:39.22*** join/#htc-linux kimhoon (n=kimhoon@s559116c1.adsl.wanadoo.nl) [NETSPLIT VICTIM]
11:39.22*** join/#htc-linux imfloflo (n=imfloflo@cap31-6-88-180-73-121.fbx.proxad.net) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux kilian_ (n=kilian@92.66.94.81) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux iNFEKT (n=nFKT@76.20.185.254) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux timebomb (n=tb@e176126245.adsl.alicedsl.de) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux BabelO_ (n=fcr@lun34-2-82-238-28-28.fbx.proxad.net) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux Zinbolic (n=Zy@84.238.80.215) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux paulproteus (n=paulprot@2002:db69:2513:0:0:0:0:1) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux pichurri (n=pichurri@users3.ilo.org) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux Nullbyte (i=nullbyte@85.17.58.168) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux rmoravcik (n=rmoravci@gtsgw.ttc.cz) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux tekkdrone (n=tekkdron@72.183.115.231) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux rl2000 (n=rl2000@0x573b8b62.hhnqu1.dynamic.dsl.tele.dk) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux ecze (n=ecze@eczema.ecze.com) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux kiozen (n=oeichler@p5492A377.dip0.t-ipconnect.de) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux jerry507_mac (n=jerry507@12-226-245-130.client.mchsi.com) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux zule (i=foglem@shadowmite.com) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux goxboxlive (n=goxboxli@mail2.hjellnesconsult.no) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux mickey (n=mickey@e180159077.adsl.alicedsl.de) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux dariball (n=dariball@p4FDC4548.dip.t-dialin.net) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux Kuma (n=John@g227196088.adsl.alicedsl.de) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux divinebovine (n=bite@S01060016b6b53675.vf.shawcable.net)
11:39.23*** join/#htc-linux br1ck (n=br1ck@xdslcy254.osnanet.de) [NETSPLIT VICTIM]
11:39.23*** join/#htc-linux DracoDan (n=F@pool-71-163-138-224.washdc.east.verizon.net) [NETSPLIT VICTIM]
11:39.24*** join/#htc-linux Tinyboom (n=nahh@108.84-49-166.nextgentel.com)
11:39.25*** join/#htc-linux ali1234 (n=al@robotfuzz.co.uk) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux Shadow____X (n=Shadow__@c-24-0-8-231.hsd1.nj.comcast.net)
11:39.25*** join/#htc-linux StarLite (n=nnscript@s55916ca6.adsl.wanadoo.nl) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux chab7 (n=kvirc@212.92.4.114) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux camden-away (n=camden@ancnat2.apto.aptalaska.net) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux Spion (n=spion@unaffiliated/spion) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux skodde (n=skodde@unaffiliated/skodde) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux Czarnas (n=czarnas@imik.wip.pw.edu.pl) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux PoohbaLT (n=Poohba@c-98-235-66-242.hsd1.nj.comcast.net) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux workkaze_ (n=kaze@ABordeaux-152-1-5-4.w82-125.abo.wanadoo.fr) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux maejrep (n=madcoder@c-71-225-238-170.hsd1.pa.comcast.net) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux sietse (n=sietse@vogons.xs4all.nl) [NETSPLIT VICTIM]
11:39.25*** mode/#htc-linux [+o ChanServ] by irc.freenode.net
11:39.25*** join/#htc-linux Miek (n=mike@unaffiliated/mikechml) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux Skitzo__ (n=DCLXVI@eth582.vic.adsl.internode.on.net) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux the_sys0p (n=the_sys0@cpe-75-85-249-111.bak.res.rr.com) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux axe_hm (n=axe_fn@cybernet.net.ru) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux Marajin (n=marajin@87-194-102-189.bethere.co.uk) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux parmaster (i=par@69.60.134.179)
11:39.25*** join/#htc-linux ALoGeNo (n=alogeno@245.Red-79-153-25.staticIP.rima-tde.net)
11:39.25*** join/#htc-linux zhobbs (n=zach@cpe-75-186-147-164.woh.res.rr.com) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux errf323 (n=david@bsdguru.net) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux thinko (n=jdoe6alp@smaug.rackdragon.com) [NETSPLIT VICTIM]
11:39.25*** join/#htc-linux toi (n=pleemans@d5153128F.access.telenet.be)
11:39.26*** join/#htc-linux dase (i=dase@p1atin.de) [NETSPLIT VICTIM]
11:39.26*** join/#htc-linux tote (i=me@c-2abd72d5.018-24-73746f48.cust.bredbandsbolaget.se) [NETSPLIT VICTIM]
11:39.26*** join/#htc-linux wasabi (n=wasabi@ubuntu/member/wasabi) [NETSPLIT VICTIM]
11:39.26*** join/#htc-linux Abracadabra (n=aaabraca@unaffiliated/abracadabra) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux pigeon (n=pigeon@60-241-137-179.static.tpgi.com.au)
11:39.27*** join/#htc-linux NAiL (n=repvik@nslu2-linux/repvik)
11:39.27*** join/#htc-linux sleepr (n=sleepr@c-9ddce355.03-113-73766c10.cust.bredbandsbolaget.se) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux j0b0 (n=jobo@5ED40048.cable.ziggo.nl) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux nizox (n=none@2a01:e35:8a13:a2b0:21c:c0ff:fe25:ff68)
11:39.27*** join/#htc-linux shoragan (n=shoragan@debian/developer/shoragan)
11:39.27*** join/#htc-linux hlbot (n=adm@iclem.net) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux era- (n=im@70.16.65.252) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux Kensan (n=ken@gw.ptr-80-238-181-161.customer.ch.netstream.com) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux NetRipper (n=netrippe@netripper.nl) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux mrmg (n=mrmg@spc1-bror4-0-0-cust383.asfd.broadband.ntl.com) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux ponicg_ (n=ponicg@ool-182d907a.dyn.optonline.net) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux dream_kill (n=nospam@92.56.48.66) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux RZK333 (n=rzk@daemonet.ru) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux tmzt (n=tmzt@76.211.12.137) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux lupine_85 (n=lupine_8@work.lupine.me.uk) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux mooky (n=suzannet@unaffiliated/mooky) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux hairyraven (n=nobody@loy.pp.ru) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux TeringTu1y (n=maarten@195-241-125-243.ip.telfort.nl) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux ImCoKeMaN (n=imcokema@pool-98-111-118-30.hrbgpa.fios.verizon.net) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux swetland (n=swetland@nat/google/x-ac2328226c12fbc1) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux Funklord (n=cow@c-ecd572d5.014-46-73746f28.cust.bredbandsbolaget.se) [NETSPLIT VICTIM]
11:39.27*** join/#htc-linux Dinde (n=kayser@81-65-176-209.rev.numericable.fr) [NETSPLIT VICTIM]
11:39.28*** join/#htc-linux tcccp (n=hey@223.66.238.89.arpa-addr.in) [NETSPLIT VICTIM]
11:39.28*** join/#htc-linux Descention (n=munsco95@acm.pct.edu) [NETSPLIT VICTIM]
11:39.28*** join/#htc-linux rakeem (n=rak@n219079053137.netvigator.com) [NETSPLIT VICTIM]
11:39.28*** join/#htc-linux lama (i=lama@netbsd.pl) [NETSPLIT VICTIM]
11:39.29*** join/#htc-linux toer (i=tore@179.81-166-86.customer.lyse.net) [NETSPLIT VICTIM]
11:39.29*** join/#htc-linux infernix (i=nix@unaffiliated/infernix) [NETSPLIT VICTIM]
11:39.29*** join/#htc-linux Kevin2 (n=Kevin2@207-237-194-161.c3-0.avec-ubr2.nyr-avec.ny.cable.rcn.com) [NETSPLIT VICTIM]
11:39.29*** join/#htc-linux nextime (n=nextime@unaffiliated/nextime) [NETSPLIT VICTIM]
11:39.29*** join/#htc-linux defiant (n=add@p0pc0rn.eu) [NETSPLIT VICTIM]
11:39.29*** join/#htc-linux iratsu (n=iratsu@modemcable018.234-59-74.mc.videotron.ca) [NETSPLIT VICTIM]
11:40.55*** part/#htc-linux rmoravcik (n=rmoravci@gtsgw.ttc.cz)
11:40.55*** join/#htc-linux balsat1 (n=kll@87.72.13.87)
11:40.58lupine_85not android. angstrom
11:40.58lupine_85d'oh
11:45.11*** join/#htc-linux tsdogs (n=tsdogs@net70-17.metalit.net)
11:45.33*** join/#htc-linux kimhoon (n=kimhoon@s559116c1.adsl.wanadoo.nl)
11:45.55*** join/#htc-linux yoyey (n=yoann@bro69-3-82-237-160-83.fbx.proxad.net)
11:48.36ALoGeNohi
11:49.23ALoGeNodcordes_ [15:18] #htc-linux <dcordes_> ALoGeNo can you help with acx patches for passive scanning?
11:49.27ALoGeNo??¿¿
12:06.50lupine_85half built :D
12:08.32dcordes_lupine_85 you using org.oe.dev e
12:08.32dcordes_?
12:09.26lupine_85dcordes_: yep
12:09.42dcordes_lupine_85 if you want modem support, build fso-console-image
12:10.01jerry507_macWhat continues to baffle me is why a single company like htc makes so many changes to each model
12:10.03lupine_85only change I've needed to make so far is a few tweaks to the linux-msm7xxx_git.bb
12:10.12dcordes_yea
12:10.20dcordes_updated srcrev?
12:10.25lupine_85yep
12:10.35lupine_85and changed the defconfig to the htcraphael one
12:10.41jerry507_macYou'd think that this just adds additional development time to each model for things that really don't need to change
12:10.55dcordes_lupine_85 the defconfigs are in already
12:10.58lupine_85jerry507_mac: the lower the cost per unit, the better
12:11.03dcordes_I can update it later
12:11.35lupine_85dcordes_: I just copied over the latest from linux-msm-2.6.25 git
12:11.42lupine_85in case anything had changed
12:11.45jerry507_macYea, I get that incentive
12:11.54lupine_85jerry507_mac: development time is cheap by comparison
12:12.07jerry507_macBut then why not develop something cheap at the start?
12:12.29lupine_85because if you can find cheaper hardware for a new model, you're going to use it
12:12.30jerry507_macIt seems they switch from product to product, from things they've done even in the very recent past
12:13.10lupine_85jerry507_mac: I work in a hosting company and we change the hardware we use for a number of reasons
12:13.14lupine_85supplier availability is another
12:13.24lupine_85every time we change, it involves some development work
12:14.03dcordes_lupine_85 unfortunately not too many packages are in the armv6-novfp repos
12:14.17lupine_85dcordes_: mm, I'm not too fussed about that
12:14.24lupine_85I can always set one up :)
12:14.28lupine_85in the future
12:14.29dcordes_lupine_85 so it's better to install stuff your need directly
12:14.49lupine_85is building a huge image which the new sd code means he'll be able to mount directly
12:15.01dcordes_not needed. I wil try make them build all thes tuff
12:15.26lupine_85#angstrom seems a friendly enough place
12:15.27*** join/#htc-linux DasFx (n=John@dasfx-lptp.euronet.nl)
12:15.36lupine_85are you an hrcraph mentor ?
12:15.57dcordes_I mentor all msm machines inOE
12:17.39dream_killhi
12:17.57dream_killdcordes.. new image for me to test?
12:18.42yoyeyhello, a new video on the htc touch hd running android =>  http://de.youtube.com/watch?v=tiDqjqUjpcU
12:19.31*** join/#htc-linux rmoravcik (n=rmoravci@gtsgw.ttc.cz)
12:19.56lupine_85dcordes_: ciik
12:19.57lupine_85cool*
12:20.11dcordes_we should test fso-image (with frontend to frameworkd)
12:20.41dcordes_iirc j0b0 tried X on his raph and it worked well
12:20.46lupine_85console-image is building right now
12:21.00lupine_85but I@d imagine most of the 2K packages are reusable in that one :D
12:21.04lupine_85so hardly wasted effort
12:21.09jerry507_macSo this is kind of a dumb question but one I can't seem to find any good info on
12:21.11*** join/#htc-linux dcordes (n=dcordes@unaffiliated/dcordes)
12:21.22dcordes_lupine_85 yep
12:21.53jerry507_macDo you guys just load your images on the sd cards using a card reader or something when you're working from linux?
12:22.27lupine_85(it was meant to build overnight, but then it hit problems with the linux-msm7xxxx package caused - at least in part - by an old bitbake)
12:22.41dcordeslupine_85: notice the last two lines http://rafb.net/p/ijtKvH24.html
12:23.06dcordeslupine_85: include them in your local.conf if you build fso images so you get the latest frameworkd with msm support
12:24.30dcordeslupine_85: they should make a new stable release of bb so people get a working version
12:25.20lupine_85dcordes: well, .12 would have been ok but I was on .10
12:25.25lupine_85(intrepid package)
12:25.54dcordeslupine_85: I'm always fetching bitbake from berlios
12:25.55lupine_85now I'm using .13
12:26.27lupine_85I should pick up another SD card purely for linux development, I guess
12:27.06dcordesyou can always have a huge one with a rich fat partition
12:27.12dcordesso you can also use it in wince
12:27.36lupine_85I have a 4G one at the moment but it's mostly full of music
12:27.40dcordesfor the kais I have a 4gb now with 1gb fat where I keep haret and the rest ext2
12:31.09jerry507_macI've got a 2GB that never got used on the phone
12:31.28jerry507_macI'm just trying to get something decent up and running but I have no card reader
12:31.32jerry507_macStupid mac
12:31.36lupine_85I'll probably grab an 8G and use that for music in the future
12:32.06dcordeslights a candle for broken 8gb class 6 uSD
12:32.26dream_kill:P
12:32.42jerry507_macAt least they're not terribly pricy
12:32.42lupine_85sounds promising
12:33.16lupine_85ho de hum, I think I shall need to build a linux/amd64 -> wince/arm toolchain too
12:33.28lupine_85unless the i586 one plays well...
12:37.57dcordesNetRipper: the defconfigs still have both machine types ?
12:41.10lupine_85yay, is packaging
12:41.37lupine_851820 / 2020
12:41.42lupine_852090 rather
12:41.51lupine_85not that I can install it until tonight
12:42.12dcordesmake sure you set the framework versioning settings in local.conf
12:52.22*** join/#htc-linux rmoravcik (n=rmoravci@gtsgw.ttc.cz)
12:56.12dcordespeople will wonder why the OE defconfigs have this
12:56.14dcordes<PROTECTED>
12:56.14dcordes+CONFIG_MACH_HTCDIAMOND=y
12:56.15dcordes<PROTECTED>
13:02.47NetRipperyou mentioned it before
13:02.57NetRipperi'll remove it after i fix the ifdefs
13:04.26NetRipperthough i still believe it may cause confusion, as people building from git will create kernels specifically for one device.. while it's now common that the kernel works on both devices
13:04.39NetRipperbut i get the point about a defconfig for raphael, including the diamond ;)
13:07.16dcordesand vice versa.
13:07.44NetRipperyes
13:08.38dcordes#if defined(CONFIG_MACH_HTCRAPHAEL) || (defined(CONFIG_MACH_HTCDIAMOND)
13:10.23NetRipperminus that ( in front of defined, yes
13:10.48StarLiteMobileor add an extra ) at the end ;)
13:13.07NetRipperdcordes, what's the difference between #ifdef CONFIG_MACH_HTCRAPHAEL and #if defined(CONFIG_MACH_HTCRAPHAEL)?
13:14.01dcordesI don't know if it's also possible with #ifdef, but with #if defined you can specify more things
13:14.09dcordeswith the || like above
13:14.42NetRipperaha, because #ifdef is also used currently
13:14.50NetRipperbut i think you can also do || with ifdef
13:15.13NetRipperi'll check what's more common in the kernel ;)
13:16.04dcordes"i'll remove it after i fix the ifdefs" so I wanted to show how I would probably do this
13:16.11dcordesdon't know what's the best way
13:16.47NetRipperyou're right, nowhere in the kernel is #ifdef used with ||, so probably doesnt work ;)
13:16.55NetRipperwhenever || is used, it's #if defined
13:16.56NetRipper:)
13:17.06dcordes:)
13:17.36NetRipperanyway i'm going to do some patch committing tonight when im at home again
13:17.51NetRipperthen i'll do this too
13:18.08dcordeswhat are you planning to commit?
13:18.18dcordeslavender.t has new improvements?
13:18.40dcordesneed to catch up with the diam thread
13:18.42NetRippermaejrep has a fixed proc_comm
13:18.54NetRipperj0b0 has a keypad driver
13:19.21NetRipperand still need to commit j0b0's draggable/toggable virtual keyboard
13:19.47NetRipperand fix this defconfig/ifdef stuff
13:20.28NetRipperdcordes not everything is on the diam thread, you can find it in irc log though
13:20.35NetRipperxda was down for a long time
13:22.58*** join/#htc-linux piusvelte (n=chatzill@nat.philau.edu)
13:28.53*** join/#htc-linux Kensan_ (n=ken@gw.ptr-80-238-181-161.customer.ch.netstream.com)
13:31.34MullinsI can use 'git checkout -b htc-msm origin/htc-msm' to get a remote branch 'htc-msm', but how can I find out the remote branches available if I didn't know them in advance?
13:34.11dcordesNetRipper: did you get the query?
13:34.23dcordesMullins: git-branch -a
13:34.39Mullinsexcellent, thanks
13:35.35*** join/#htc-linux imfloflo (i=5bcfd11d@gateway/web/ajax/mibbit.com/x-954987cd2bb0ccb5)
13:36.56dcordesimfloflo: did you try the new polaris kernel?
13:37.20*** join/#htc-linux pleemans (n=toi@d54C2AAB7.access.telenet.be)
13:40.19imfloflonow
13:40.28imfloflonot now
13:40.32imfloflowhere is it ?
13:40.44imfloflodzo kernel ?
13:40.54dcordesno. rogro kernel which is based on dzo kernel.
13:41.06dcordeshe might have posted it in the polaris thread
13:41.06imfloflojsut read http://git.linuxtogo.org/?p=groups/mobile-linux/kernel.git;a=commit;h=4726e20e839a1644626560dfa72b3e0e730ba86a
13:41.39dcordesimfloflo: oh cool :) thanks
13:42.03imfloflobut haven't see new kernel from rogro yet
13:42.19dcordesopens yet another bottle of champagne
13:42.44imfloflodzo rocks :)
13:42.49dcordesyea definetly
13:43.14imfloflowhere have you see new kernel form rogro ? nothing on xda
13:43.30dcordesthe dzo commits might base on his changes
13:44.00dcordesimfloflo: what about the good old polaris linux thread? rogro didn't post it there?
13:44.36dcordeshttp://forum.xda-developers.com/showthread.php?t=396782&page=179
13:44.37imfloflothe old one ? or the new thread?
13:45.44imfloflothat for kaiser ? i have polaris
13:45.56dcordesno it's for the polaris in the .zip .
13:46.07dcordesbut it's better to try the dzo commits now
13:46.48imflofloi use f00bar version for the moment
13:47.07*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
13:47.41dcordesl0l. not the polaris-lol version?
13:47.46lupine_85*phew* scary moment at work there :D
13:48.07lupine_85you know the moment where your broken code gets pushed to production and you realise it could screw up a lot of stuff....
13:48.27dream_killhaha
13:48.30imfloflohaven't see this version where is it?
13:48.45dcordescrap I need to downgrade my kaiser radio rom
13:49.06imfloflodowngrade :o !
13:49.22lupine_85I came close to destroying the dhcp config and serial console access for every machine on our network
13:49.28lupine_85(~1K of them)
13:49.39dcordes" sd driver only works with radio 1.65.24.36."
13:49.56*** join/#htc-linux Guimli (n=guimli@ecu69-1-82-231-127-213.fbx.proxad.net)
13:50.01dcordesI never considered the AMSS as important to SD
13:50.30imfloflowhen the polaris lol version has been release ?
13:50.35dcordeslupine_85: you do admin stuff?
13:50.57dcordesimfloflo: it was when we tested some polaris kernels. rememeber?
13:51.16lupine_85dcordes: I@m a developer / support bod / sysadmin for Bytemark
13:51.24imflofloyes but its an old version f00bar is better no ?
13:51.24lupine_85little hosting company
13:51.29lupine_85I also do some rack monkeying
13:51.30Mullinsdcordes: which config file should be used with that latest Dzo commit?
13:51.41dcordesah sounds intersting / broad
13:51.51dcordesrack monkeying ^^
13:52.09lupine_85the company employs 7 people, so you have to do a bit of everything :D
13:52.46dcordesMullins: you can do "make vogue_defconfig; make menuconfig"
13:53.05dcordesand then set the kaiser relevant configs manually
13:54.09dcordesthat's what I'm gonna do now
13:56.14dcordeswe need to add the tsc2003.c for kais
13:56.21dcordesand vogue-smd depends on vogue
13:57.19lupine_85does fso-console-image
13:57.56lupine_85yay, it can reuse 900 of the tasks
14:00.59tekkdronemorning folks
14:01.06dcordesmorning tekkdrone
14:01.28tekkdronedcordes: I put together an sd last night for that frameworkd test, might be user error but I think there's a binary incompatibility
14:02.25tekkdronenot rootfs when attempting to switch_root, chroot = illegal instruction, and I get not found when attempting to execute bins from the rootfs
14:02.47dcordesdid I put a rootfs for the wrong arch again -_- ?
14:03.01tekkdronelol maybe, or it could be user error, it was late
14:03.06tekkdrone>_<
14:03.07dcordesI was building for blueangel iirc which I thought has same arch as hima
14:05.32*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
14:05.59tekkdroneI'm using the same initrd and kernel I used when originally working on android m5 with some changes to the initrd to mount p2 and switch_root
14:06.23tekkdronedidn't have any luck with .21 or .25 (saw some .25 boot files in your image, thought I'd give my .25 kernel a shot)
14:08.26tekkdronefwiw I'm using emdebian's tools (gcc 4.2.4) against libc 2.7
14:08.27dcordestekkdrone: what are you going to do? android? chroot? angstrom?
14:08.35dcordes*_*
14:08.56dcordesKensan: thanks
14:09.14Kensandcordes: I assume you don't need the kernel
14:09.32dcordescorrect
14:10.21Kensandcordes: just tell me if you want me to rebuild etc. Since I've done the build once it should be fairly quick.
14:10.28lupine_85there's a point. is there any particular reason why 2.6.25 is the branch being worked against? any particular (dis)advantages to rebasing on the latest tree?
14:11.05tekkdroneI've moved to .27, it's advantagious developing for the Apache to have the mainline pxa27x-udc and htc-egpio modules
14:12.07tekkdroneusbserial is already working in far less time than it took me to get the hh20 pxa27x-udc module working in 2.6.25 with android ADB
14:12.12dcordeslupine_85: I don't see why we shouldn't try. tmzt mentioned 2.6.25 -> 2.6.27 should not be major
14:12.40tekkdronemost of my reasons are arch-specific though... so YMMV
14:13.12dcordeslupine_85: also the android-msm-htc-2.6.25 which is the base for htc-msm-2.6.25 and the newer google branches are for the trout
14:13.15dcordesso
14:13.30tekkdroneI do know the PM code got significantly revamped in the 2.6.27 kernel and userspace
14:14.16tekkdronecoming from my co-worker with a G1, they're eating batteries like an Apache with WiFi on
14:14.27dcordeslupine_85: I'm keen to help. my previous attempts of kaiser rebases ended in nothing but headache though
14:14.56*** join/#htc-linux danny079 (n=danny@89.146.26.66)
14:15.36lupine_85dcordes: I can certainly add it to the list of tyhings to try :D
14:15.50lupine_85(also the list of things I have little experience in.. but hows better to learn?)
14:18.12tekkdronedcordes: two questions since I can't get linuxtogo.org to load, is it there that the angstrom repo is maintained? and 2) is the frameworkd in repository as well?
14:18.38*** join/#htc-linux exco (n=exco@e181122049.adsl.alicedsl.de)
14:19.26*** join/#htc-linux marmotta (n=skodde@unaffiliated/skodde)
14:19.38lupine_85also, given the current less-than-enviable state of the repositories we work with, I'm vaguely considering grabbing myself a dedicated server and setting up a mirrored git/everything we need
14:19.49lupine_85would people be interested in hasing?
14:20.11tekkdronedid you just lolspeak?
14:20.15tekkdronelol
14:20.18lupine_85I do it from time to time
14:20.21dcordestekkdrone: yes. angstrom is a distro built from the openembedded metadata. the org.openembedded.dev branch is in the linuxtogo git. I always refer to this for quickstarting with oe angstrom-distribution.org/building-angstrom
14:20.22lupine_85it always can has lol
14:20.33dcordestekkdrone: frameworkd is hosted at git.freesmartphone.org
14:21.08tekkdronedcordes: thanks, I'd never used the angstrom distro before unpacking your rootfs
14:21.21tekkdronemy pre-android linux is LFS
14:21.39dcordesangstrom is the bestest embedded linux
14:21.52lupine_85used another one some time ago
14:21.57tekkdronenoted, I'll grab and build in the near future :)
14:21.59lupine_85erm. AspisOS
14:22.07lupine_85that appears to have sunk without a trace though
14:23.21dcordesOE eats all embedded linux
14:23.30tekkdroneFunny, same thing seemed to happen to development on Apache sometime in early 2007
14:23.48tekkdrone@poof@@
14:25.12dcordestekkdrone: btw is your apache stuff publicly accessible?
14:25.40tekkdroneit should be by this weekend, kernel was a moving target until last weekend when I settled on .27
14:26.07tekkdronegot that booting w/keyboard support, working on bringing in egpio support, keeping blowing up the kernel though
14:26.15tekkdrone>_<
14:27.12tekkdronedcordes: binaries yes but before last weekend the source was something I wouldn't give to an enemy, nevermind a colleague
14:28.46*** join/#htc-linux radem205 (n=aaa@92-108-47-154.dynamic.upc.nl)
14:29.06lupine_85hehe
14:29.10lupine_85hates looking at his old code
14:29.18lupine_85funnily enough, I'm all about that today
14:31.00tekkdroneplus, since all my internal shite is SVN I need to migrate to git before going live on the net since git seems to be preferred for kernel dev
14:34.22tekkdroneoff to work for me, talk you all shortly
14:35.26*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
14:37.28Mullinsdcordes: using tsc2003.c and kaiser-smd.c now but get an error conflicting types for 'msm_proc_comm'. Missing a config option am I?
14:39.38Mullinsdcordes: Also there is a comment in vogue-smd.c that Dzo appears to use this for Kaiser as well as Vogue. He has a condition for non-vogue devices
14:39.49*** join/#htc-linux kilian_ (n=kilian@92.66.94.81)
14:48.00dcordesMullins: so we just build both configs?
14:50.35Mullinsdcordes: compiles no problem with vogue_defconfig . it starts booting on kaiser but fails. Cant get the kaiser config to compile myself yet.
14:53.20*** part/#htc-linux danny079 (n=danny@89.146.26.66)
14:54.05*** join/#htc-linux AstainHellbring (n=AstainHe@unaffiliated/astainhellbring)
14:55.58lupine_85ho hum, build of connman is broken
14:57.42dcordeslupine_85: lol that's still not fixed?
14:57.56dcordeslupine_85: remove the connman git recipe
14:59.23*** join/#htc-linux br1ck (n=br1ck@xdslcy254.osnanet.de)
15:03.20lupine_85dcordes: hehe, done
15:03.28lupine_85crazy brokennesses
15:03.54*** join/#htc-linux orux (n=jose@89.130.46.3)
15:04.46*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
15:06.41*** join/#htc-linux timebomb (n=tb@e176098059.adsl.alicedsl.de)
15:08.46*** join/#htc-linux Heema (n=weechat@217.139.57.79)
15:09.25lupine_85ho hum, and 'ERROR: QA Issue: libpanel.la failed sanity test (workdir) in path /opt/OE/angstrom-stable//staging/armv6-novfp-angstrom-linux-gnueabi/usr/lib
15:09.35lupine_85(and for a couple of other .la s)
15:10.53*** part/#htc-linux Heema (n=weechat@217.139.57.79)
15:12.23dcordeslupine_85: didn't get those
15:16.47lupine_85deletes
15:18.00*** join/#htc-linux GPFerror (n=gpferror@cpe-76-187-41-132.tx.res.rr.com)
15:18.56lupine_85that seems to have done the trick
15:19.28*** join/#htc-linux MethoS- (n=clemens@host-091-097-243-110.ewe-ip-backbone.de)
15:23.08*** join/#htc-linux tsdogs (n=tsdogs@net70-17.metalit.net)
15:23.17lupine_85yep, I have a rootfs directory :D
15:24.42*** join/#htc-linux cr2 (n=konversa@crpl6.physik.uni-wuppertal.de)
15:24.47lupine_85Angstrom-fso-console-image-glibc-ipk-2009.X-test-20090107-htcraphael.rootfs.ext2
15:24.50lupine_85woohoo
15:26.27*** join/#htc-linux tekkdrone (n=none@72.183.115.231)
15:27.52AstainHellbringhiya lupine_85 whatcha breaking today?
15:29.05lupine_85AstainHellbring: right now - nowt
15:29.12lupine_85when I get home - anything that feels like it
15:29.34lupine_85sets an x11-image off as well
15:29.36AstainHellbringfun fun
15:30.06lupine_85'twill be if it works :D
15:32.38lupine_85oof,  652 of 3884
15:34.18tekkdronemight be a while
15:34.42lupine_85possibly
15:34.47lupine_85(768 now)
15:35.12lupine_85about half of them are simple tasks by the looks of it
15:35.15*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
15:35.16lupine_85the other half being compilations
15:35.36tekkdronehmmm, looks like this openembedded is something I should look into
15:35.47tekkdronespeaking of compilations
15:35.59lupine_85it is remarkably nice
15:36.18tekkdroneI've been kludging through with emdebian's cross tools
15:36.42dcordesbut it's also a rapidly moving target
15:37.25tekkdroneit appears also to be the build environment of choice for people working with angstrom
15:37.46lupine_85I just like things that automate builds
15:37.55lupine_85hates setting up build environments
15:38.30dcordesgoogle should have used bitbake as well
15:38.49cr2hi
15:38.52lupine_85google should do lots of things :D
15:39.00cr2i'm loking for somebody with msvc
15:39.12dcordeshi cr2
15:39.30cr2who can crosscompile a function call with 7 parameters
15:39.31tekkdronewhich version cr2?
15:39.34cr2hi d cordes
15:40.01cr2tekkdrone: latest ?
15:40.17tekkdroneI can build for x86 targets here but my EVC stuff is at home
15:40.17NetRippercr2 cegcc is not sufficient?
15:40.23cr2tekkdrone: the one htc used to compile the ms m720xA dlls
15:40.43cr2NetRipper: i'm interested in the msvc stack layout management
15:40.46dcordeslupine_85: btw, the linux-msm7xxxx srcrevs and defconfigs were updated in org.oe.dev
15:40.48NetRipperah
15:41.00lupine_85dcordes: I spotted it in the update :)
15:41.16tekkdronecr2: I've got vs2008 with all the mobile goodies at the house and can do it at lunch if you haven't by then
15:41.56cr2NetRipper: raph500 is 7500A ? it's just cosmetic difference. i thin ki've doumented the way to detect 1/0A
15:42.14cr2tekkdrone: thanks
15:42.33tekkdroneit's 9:42 local time, lunch is noon for me
15:42.52cr2hehe, 16:42 here
15:43.05tekkdroneNetRipper: did maejrep by anychance pass along a build of your kernel that skiped FB handoff?
15:43.08tekkdroneskipped*
15:43.12dcordescr2: I added HTCRAPHAEL-CDMA and HTCDIAMOND-CDMA machine types
15:43.23AstainHellbringcr2 yes raph500 is 7500a not 7501a
15:44.14cr2AstainHellbring: are there any real differences between the cpus? somehow i doubt it
15:44.40cr2afair it was something about some weird patent war
15:44.54tekkdronewith broadcom, I think
15:45.01cr2yes
15:45.12AstainHellbringyes ISM memory
15:45.35AstainHellbring7500 proc has 64mb ism 7501a has 32mb ism
15:46.08cr2AstainHellbring: ? my 7201A uses 64mb sram
15:46.44cr2nonA is 32 and A has 64
15:47.34cr2dcordes: what about alt gpio msm config ?
15:47.40tekkdroneapparently there are some other differences as well with the raph500,
15:47.51cr2dcordes: i see some alt differences for raph800 in the wiki
15:48.15NetRippercr2 im not sure, someone posted haret info string of the raph500, you can find it in logs somewhere :)
15:48.16tekkdronee.g. we were testing one of netrippers kernels on my raph500 last night and when the console handoff occurs on-screen output stops
15:48.20cr2tekkdrone: split the wiki description coulmn into 3. i can't edit wikii
15:48.37NetRippercr2 oh astain alreayd answered you :)
15:49.06NetRippertekkdrone you can disable the handover with a .config option
15:49.18tekkdroneaye, maej was going to do that
15:49.23tekkdroneI don't have the source
15:49.30cr2btw, the LCD-related vreg stuff can be managed through proc_comm_wince now ?
15:49.34NetRipperah
15:49.46tekkdronehe was thinking panel differences, and the usb wasn't working either (it's my understanding it was working for raph800)
15:50.04NetRippercr2, what lcd related vreg stuff?
15:50.29cr2NetRipper: there are 2 calls in the wiki
15:50.40NetRippertekkdrone, sec.
15:51.05tekkdroneNetRipper: no problem, I'm at work, not going anywhere till lunch but the side door for a cigarette ;)
15:51.06NetRippercr2 the pmic lcd on/off?
15:51.09cr2g1/ndroid code created some obfuscated mess out of the clock/name issue
15:51.26cr2not only the lcd, maybe the mddi too
15:51.57NetRippercr2, what is pmic?
15:52.13cr2we also have an issue with the 2 param proc_comm_wince calls
15:52.15NetRipperand can it be used other than just turning lcd on/off?
15:52.31NetRipperproc_comm has 2 data params
15:52.38NetRippersame as the proc_comm.c
15:52.52cr2it needs better docs, and better wince ioctl compatibility. i'll try to look at it in the coming days.
15:53.24cr2NetRipper: proc_comm_wince is implemented as ioctl(DEX:,...)
15:54.01cr2don't mix g1 proc_comm and wince "proc_comm"
15:55.16NetRipperah ok
15:55.21NetRipperhm
15:55.41*** join/#htc-linux daspsycho (n=Tim@HSI-KBW-082-212-005-059.hsi.kabelbw.de)
15:55.57cr2if proc_comm_wince will be compatible with the ioctl syntax, it'll be much easier to trace on new devices, and document things
15:56.26NetRippercr2, so it's not a good idea to just have a proc_comm_wince.c that has the same interface (i.e. function prototypes) and perform changes in Makefile?
15:57.00NetRippercurrently it already has the same function prototypes
15:57.05NetRipperthis is how i set it up locally
15:57.27cr2it's ok when it works, but imho it creates a lot of misunderstanding.
15:58.10cr2for the people who don't know about the amss ipc differences
15:58.16NetRipperit works.. i did it like that because it would save a lot of ifdefs all around the code for each current proc_comm call
15:58.45NetRipperi didn't notice that it had differences like this
15:59.31NetRippertekkdrone, boot with this kernel, see what happens: http://www.netripper.com/raphael/zImage-raph500
15:59.57NetRippertekkdrone, it has some experimental stuff in it with some extra code, but at least it wont do the handover
16:00.01*** join/#htc-linux Heema (n=weechat@217.139.57.79)
16:00.15cr21c   0x4   pmic vibra on
16:00.18NetRippertekkdrone, im not sure how to find out the id of your mddi_panel, but maejrep probably knows
16:00.30cr240   0x17   vibra motor on
16:00.40NetRippervibra works
16:00.43cr2NetRipper: can you explain the difference ?
16:00.49NetRipperno, i dont know what pmic is
16:01.08NetRipperthe 0x17 one works though and 0x18 to turn it off
16:01.18cr2NetRipper: dzo used 0x4, not 0x17 in his vogue code
16:01.27NetRipperi can try 0x4 as well
16:01.31cr20x4/0x5
16:01.45NetRipperyou dont have any clue what pmic might be?
16:02.08cr2PowerManagementIntegratedCircuit is pmic
16:02.23NetRipperah
16:02.47cr2it's controlled over SBI (it's something like w1 "innovated" by qualcomm)
16:03.12*** join/#htc-linux Abracadabra (n=aaabraca@62-244-191-249.cust.exponential-e.net)
16:03.19cr2which resides somewhere in the deep dark microkernel internals
16:03.31cr2which g1/android/htc does not want to tell you about
16:03.31NetRipperso if that works, there would be two ways to control the same thing... which makes me wonder what the right one is ;)
16:04.04cr2there are 2 ways to access it: DEX ioctl, and the magic hidden AT cmd channel
16:04.34NetRipperyes but two ioctl's for the same thing would be confusing
16:05.07cr2dzo traced 1f SMEM area
16:05.25cr2so vogue uses 0x4/0x5
16:05.25*** join/#htc-linux dcordes_ (n=dcordes_@unaffiliated/dcordes)
16:05.36cr2you may try to check raph100
16:05.56cr2does the "nled" control vibra on raph ?
16:06.00*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
16:06.45NetRipperyep 0x4 and 0x5 work as well
16:06.46NetRipperso what should we use, 0x4/5 or 0x17/18
16:07.31cr20x4/0x5 for better compatibility
16:07.54cr2but these are "void" functions
16:08.09NetRipperno params needed you mean?
16:08.19cr2we need to check the functions transmitting  paramter(s)
16:08.22cr2yes
16:08.33NetRippermaejrep was tracing one i think
16:08.37NetRipperit was in log somewhere
16:08.38NetRipper:)
16:09.03cr2which i've documented as 2=0xX
16:09.20cr2confusing too ;)
16:09.48NetRippercr2:
16:09.50NetRipper04:42:49 < maejrep> http://privatepaste.com/c37MKMwzUQ <-- some PCOM command that runs every 10s
16:09.53NetRipper04:43:03 < maejrep> http://privatepaste.com/080PloxouC  <-- something related to inserting the SD card into the slot
16:09.56NetRipper04:46:09 < maejrep> a few differences:  1) the SD command requires data, the other doesn't.  2) SD command re-sets the cmd registe and the other doesn't, 3) Proc sets status register equal to the CMD value in the SD command, but in the other 0x100 | CMD
16:10.00NetRipper04:46:52 < maejrep> I'm thinking the 0x100 bit in COMMAND tells the processor to expect data in DATA1.  and 0x100 in STATUS regist that it should expect data in DATA_RESULT1 ?
16:10.03cr2ok, so what vreg* was used for lcd
16:10.04NetRipperyou can read the rest in log
16:10.06NetRipper;)
16:10.21cr2NetRipper: these are  batter status/id queries
16:10.41cr2just some timer polling
16:11.01cr2imho the most complex is
16:11.05cr248   0=0x1a   pmic reg * set voltage *
16:11.43cr2whicxh is a "generic" pmic reg+voltage control
16:12.15cr2if he can trace its ABI that'll be great
16:12.45NetRipperhow about set audio path.. if i look at the audio paths config in vogue it seems too much to pass in 2 params :)
16:13.30cr244   2=0x19   set audio path
16:13.30NetRipperwhat does ABI stand for?
16:13.40NetRipperyes.
16:13.45cr2BinaryInterface
16:13.50NetRipperok
16:14.15NetRipperthink im gonna start an acronyms page ;)
16:14.21cr2what you put in which register/stack on a "call"
16:15.00cr2it's a must to know, if you do RE
16:15.57NetRippertracing abi with haret means putting a trace on the r0-r11 registers then?
16:16.25cr2no. there are some ABI conventions
16:16.36cr2which are compiler-specific (OS specific too)
16:16.56NetRipperoh you mean the order in which parameters are pushed upon the stack and that kind of thing?
16:17.07cr2that's why i'd like to see a 7 param call done by msvc
16:17.29cr2because 4-5 params are put into the registers
16:17.45cr2but for 7 you need to setup a stack fram
16:18.05cr2afair , it's all came from dark unix ages :)
16:18.16cr2and the PDP cpu
16:18.52NetRipperah, right
16:19.28cr2evev if you have 32(31) register like on alpha, the ABI tells you to use only the first 5 (afair)
16:20.08cr2so it's just a convention
16:20.18cr2important only for the syscalls, and RE people :)
16:20.56NetRipperhehe
16:21.07NetRipperwhat does the * denote in the DEX overview?
16:21.58cr2NetRipper: do you check "XC" for the lcd ?
16:21.58cr2check PCB XC
16:21.58cr2if (XC) gpio_0x63=1
16:21.59cr2else    gpio_0x52=1
16:22.06NetRipperthat it expects a memory address to a function?
16:22.10cr2* is some value here (param)
16:22.22NetRipperok
16:22.31cr2no, a register value
16:22.41NetRipperoh in that case, what does the 2= in front of the 0x19 mean?
16:22.46cr2may be a pointer too
16:22.48NetRipperi thought that was the param ;)
16:23.00cr2it all depends on the ioctl ABI
16:23.34cr2hmm, check the wince ioctl prototype
16:23.46cr2deviocontrol
16:24.28cr2vreg(4,1,bb8) : vreg_aux ?
16:24.28cr2vreg(b,1,708) : vreg_rftx ?=bit 0x800 ?
16:25.11cr2we need to find out which proc_comm ioctl number/cmd corresponds to that.
16:25.33cr2evil qualcomm ;-)
16:25.45*** join/#htc-linux tsdogs (n=tsdogs@net70-17.metalit.net)
16:27.29NetRipperhm
16:27.42cr2if the lcd driver calls DEX ioctl, it should be visible
16:29.29NetRipperwe can trace proc_comm addresses while doing lcdpower on/off
16:30.16*** join/#htc-linux kiozen (n=oeichler@rgnb-5d87c0bd.pool.einsundeins.de)
16:33.58*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
16:35.04NetRipperim off, back to home ;)
16:35.12NetRipperreally productive workday i had..
16:37.33tekkdronelol
16:42.22tekkdronehmm, panic on ripper's kernel with raph500
16:50.25*** join/#htc-linux Shadow____X (n=Shadow__@c-24-0-8-231.hsd1.nj.comcast.net)
16:54.51*** join/#htc-linux Guimli (n=guimli@ecu69-1-82-231-127-213.fbx.proxad.net)
16:59.02*** join/#htc-linux radem205 (n=aaa@92-108-47-154.dynamic.upc.nl)
17:04.30*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
17:08.12*** join/#htc-linux stefan_schmidt (n=stefan@p5B035756.dip.t-dialin.net)
17:11.37*** join/#htc-linux myxor (n=myxor@pdbn-4d089087.pool.mediaWays.net)
17:12.11*** join/#htc-linux pleemans (n=toi@d54C2AAB7.access.telenet.be)
17:13.51lupine_85copies his spanking new images to t'internet - and fro/ there hopefully to my phone
17:14.27lupine_85turns out i can has 500mb of internets on my price plan :)
17:15.08*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
17:15.51lupine_85doing this kind of thing on the train = lol
17:25.37tekkdroneahhh mobility :)
17:29.27*** join/#htc-linux maejrep[w] (n=madCoder@smtp-n.myyearbook.com)
17:30.34*** join/#htc-linux rmoravcik (n=rmoravci@gtsgw.ttc.cz)
17:36.25*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
17:51.17*** join/#htc-linux goxboxlive (n=goxboxli@24.84-48-212.nextgentel.com)
17:59.28*** join/#htc-linux Heema (n=weechat@217.139.57.79)
18:03.41*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
18:05.37tekkdronecr2: on my way home for lunch if you still need a compile, I'll talk to you from there
18:12.53*** join/#htc-linux stefan_schmidt (n=stefan@p5B035756.dip.t-dialin.net)
18:13.09*** join/#htc-linux pichurri (n=pichurri@194.230.146.93)
18:15.19*** join/#htc-linux balsat1 (n=kll@87.72.13.87)
18:15.23NetRippertekkdrone_, make a picture of the panic or type the first few lines.. it mentions some memory addresses it failed on
18:15.46*** join/#htc-linux rubberducky (n=rubberdu@83.101.2.232)
18:15.48rubberduckyhello
18:29.33*** join/#htc-linux tekkdrone (i=tekkdron@72.183.115.231)
18:30.15tekkdronecr2: msvc ready to go
18:32.50pichurriNetRipper,
18:32.56NetRipperthat's me
18:32.58pichurrihello NetRipper ...
18:33.16*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
18:33.34tekkdroneNetripper: that test kernel panicked out on raph 500, I'll give you the details in just a moment, booting it again
18:33.36pichurrijaja, sorry, do you know if MDDI_CLIENT_CORE_BASE could have something to do with the screen output problem for the blackstones...?
18:33.46*** join/#htc-linux Mullins07 (n=bw@89.204.247.31)
18:33.46cr2tekkdrone: create a function param7(int r0, int r1, ..., int r6)
18:34.10cr2doing something like return r0+r1+...+r6
18:34.27pichurriNetRipper, or do you have any ideas on how to debug the problem...
18:34.36cr2then call it from main, and put the binary somewhere
18:34.49cr2bbl, need to buy some food.
18:35.46cr2NetRipper: btw, i've connected to haret over usb on raph100
18:36.11cr2but it's a bit pita, because i don't have inet connection on the notebook
18:36.25tekkdronecr2: what parameters would you like passed to the param7 function?
18:36.26NetRipperpichurri, screen problem? you dont get console on blackstone?
18:37.17pichurrijust via ssh, but the screen output goes like http://www.gavero.com/pichurri/blackstone/images/corrupted_output_2008-12-28_4_small.jpg
18:37.39pichurriNetRipper, ssh via usbnet...
18:38.06NetRipperpichurri, oh,
18:38.24NetRipperpichurri, i dont have experience with those graphical glitches
18:38.38NetRipperbut this is even worse than it was on raph800
18:38.47NetRipperer, xperia i meant
18:39.03pichurriNetRipper, yep, do you know who could point me somewhere...?
18:40.01NetRipperi think j0b0 tried to get some graphical glitches working for raph800 with color packing and stuff like that.. maybe he can guide you a bit.. he'll be a bit more familiar with what is possible regarding setting parameters etc.
18:40.37pichurriok, thanx!
18:40.55pichurriby the way, congrats on the progress on the raphaels!
18:41.09*** join/#htc-linux orux (n=jose@89.130.46.3)
18:44.13tekkdroneNetRipper: here's the panic I'm seeing: Unhandled fault: external abort on linefetch (0x006) at 0x00000000
18:44.13tekkdroneModules linked in:
18:44.13tekkdroneCPU: 0Not tainted(2.6.25-00670-gc07884f-dirty #52
18:44.13tekkdronePC is at cache_alloc_refill+0x4c4/0x5e8 (also the call at the top of the backtrace)
18:44.13tekkdroneLR is at etext+0x0/0x0
18:44.40*** join/#htc-linux Zoolooc (n=fredsiba@nrbg-4dbfe2db.pool.einsundeins.de)
18:44.52*** join/#htc-linux dwaradzyn (n=chatzill@host40-58.lerkins.com)
18:45.54NetRippernice, nullpointer
18:46.33NetRipperthough the function doesn't look familiar at all.. can you list a few more functions? i'll take a quick peek when i get back
18:46.43NetRipperfrom the backtrace
18:46.46NetRippergtg, sports
18:48.06tekkdronesure thing, and it's early on in the kernel bringup, right after the Memory: N available printk
18:48.19NetRipperhm ok
18:48.30NetRipperin that case
18:48.32NetRippertry a lower mem=
18:48.35NetRippersuch as mem=32
18:48.36NetRippersuch as mem=32M
18:49.03NetRipperbbl
18:49.07tekkdronetake it easy
18:49.07dwaradzynpichurri: i tried some trial and error changes in mdp.c mdp_dma_to_mddi. the results were more corruption or lcd poweroff. no progress at all
18:52.18dwaradzynswetland: do you have any experience with fb and mddi? can you give us any suggestion about corrupted output on lcd on blackstones?
18:57.36dwaradzynbbl
19:00.46*** join/#htc-linux exco (n=exco@e181122049.adsl.alicedsl.de)
19:05.46*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
19:07.35cr2pichurri: can you dump the spl ?
19:08.06cr2pichurri: is there any wiki doc about blackstone hw setup ?
19:08.44cr2tekkdrone: 1,2,3,4,5,6,7
19:12.52tekkdronecr2: ok, last question, target device?
19:17.12*** join/#htc-linux marked (n=marked@host81-136-172-72.in-addr.btopenworld.com)
19:18.14*** join/#htc-linux imfloflo (n=imfloflo@cap31-6-88-180-73-121.fbx.proxad.net)
19:22.10*** join/#htc-linux stefan_schmidt_ (n=stefan@p5B037F1A.dip.t-dialin.net)
19:23.39cr2tekkdrone: wince6 ?
19:28.09tekkdronehmm I don't have the 6 SDK's but I have 5R2 on hand and the function is known to it, but I'm sorry what I meant was smartphone or ppc sdk, I'm guessing you're targeting raph?
19:28.37tekkdroneI can have the 6 sdk's this evening though after work
19:29.01cr2tekkdrone: sdk does not matter. compile for wince5
19:35.58*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
19:36.26*** join/#htc-linux balsat (n=kll@87.72.13.87)
19:38.15maejrep[w]trying to compare the asm via a known source? ;x
19:40.18cr2maejrep[w]: yes
19:41.30maejrep[w]I tried to reverse the original parameters to arrive at what I saw in haret, but I couldn't get values that would work
19:41.58maejrep[w]that is, working backwards from your formula to get the 7 parameters
19:42.24cr2i'm sure in the first 4
19:42.40cr2but the remaining 3 need to be checked
19:43.44cr2maejrep[w]: do you think w eshould implement the hiden AT cmd channel ?
19:44.04cr2i'd prefer the proc_comm_wince way
19:46.32*** join/#htc-linux Czarnas (n=czarnas@imik.wip.pw.edu.pl)
19:47.46cr2i've seen tht there are 2 more keyboard leds on raph
19:48.37cr20xce*2
19:49.07maejrep[w]hidden AT cmd channel being which?
19:49.11maejrep[w]is that channel 1?
19:49.26cr23
19:49.37maejrep[w]oh, i have no idea about that
19:49.45cr2but with a separate control area
19:50.01maejrep[w]channels 0 and 1 on raph800 have to be created like in vogue-smd.c
19:50.17maejrep[w]but on raph100, they follow the alloc table method as with the others
19:50.29cr2strange
19:51.03maejrep[w]and raph800 also doesn't seem to have channel 27, or have gps data at all in smem
19:51.44maejrep[w]what is the hidden AT channel for?
19:52.02maejrep[w]same things that proc_comm does?
19:52.30cr2to send pmic commands as plain text
19:53.37cr2y wiki docs say 0xce is the keyboard id
19:53.47cr2ce,12,02,x
19:54.02cr2it's probably the key read
19:54.27cr2id,addr,size <- read
19:55.09cr2it should be easy to check with i2c mmutrace
19:56.52maejrep[w]when I try to trace i2c, it freezes my phone
19:56.58maejrep[w]at least with the i2c addresses in the wiki
19:57.14maejrep[w]0xce is what?  the proc comm ioctl?
19:57.31*** join/#htc-linux lpotter (n=ljp@58.173.176.153)
19:58.19cr2no, the i2c id
20:03.32tekkdronecr2: binary file comes back with 1C 00 00 00, (hex for 28, (1+2+3+etc etc)) will this work for your needs?
20:06.08*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
20:07.40cr2i think yes
20:07.40tekkdronehttp://chillaxin.net/cr2_check.zip
20:07.40cr2ok
20:07.40tekkdronethere you go, debug is enabled on the binary if you need it
20:08.19cr2downloading
20:08.32tekkdronehttp://pastebin.com/m1006fc80 quick-and-dirty source code
20:08.41tekkdroneback to work, ttyiab :)
20:08.55cr2thanks :)
20:09.07cr2raph gps is not very good ;)
20:09.15cr2compared to sirf3
20:10.45dream_killhi guys
20:10.51*** join/#htc-linux Guimli (n=guimli@ecu69-1-82-231-127-213.fbx.proxad.net)
20:10.52dream_killany updated on raph100 ?
20:13.22Mullinsdcordes: have got Dzo's latest commit starting Android on Kaiser. One problem is that it goes to Power off straight away
20:21.24cr2hmm. disassembled
20:21.37cr2it uses the sane way
20:21.53cr2r0-r3 in the regs
20:22.08cr2the rest is sp,sp+4 and sp+8
20:22.24cr2spl is different ;-)
20:22.33cr2need to think about it.
20:23.00*** join/#htc-linux Xime (n=xime@bankize.net)
20:23.22cr2maejrep: the last 3 params are probably different for SD
20:28.31cr2stmfd sp!, {r4-r11,lr}
20:28.40cr2what happens with sp here ?
20:29.29cr29 registers stored
20:30.01cr29*4=36
20:30.11cr2=32+4
20:30.18cr2=0x24
20:30.56cr2ok, so the #0x24 is an offset for the stack frame
20:33.28cr2so the order is 18=sp,14=sp+4,10=sp+8 and c=sp+c
20:35.54*** join/#htc-linux lolwat (n=nFKT@216.163.57.57)
20:36.03*** join/#htc-linux MethoS-- (n=clemens@host-091-097-243-110.ewe-ip-backbone.de)
20:37.10*** join/#htc-linux miknix (n=miknix@gentoo/developer/miknix)
20:37.48cr248 mhz: x, 1, 0xc, 6, 1,3,1,1
20:37.59cr2does it match the old values ?
20:41.45*** join/#htc-linux Mullins (n=bw@89.204.247.31)
20:43.20*** join/#htc-linux yoyey1 (n=yoann@lns-bzn-49f-81-56-185-87.adsl.proxad.net)
20:43.50*** join/#htc-linux Mullins (n=bw@89.204.247.31)
20:46.53*** join/#htc-linux tekkdrone (n=none@72.183.115.231)
20:49.40*** join/#htc-linux Heema (n=weechat@217.139.57.79)
20:50.24tekkdroneNetripper: that kernel booted with a smaller mem specified, did you hardcode a delay into the kernel messages?
20:51.42cr2androed spl is compiled with some bizarre cmpiler
20:53.38tekkdroneon a side note, the MS SDK builds all the windows stuff against ARMV4I
20:54.28tekkdroneI'm pretty certain the ce binaries on our devices are the same, (ARMV4 or ARMV4I)
20:56.17cr2it seems i have a wrong .text offset
20:56.21*** join/#htc-linux simmel80 (n=simon@DSL01.212.114.235.167.ip-pool.NEFkom.net)
20:56.32cr2ok, emapi
20:57.15*** part/#htc-linux yoyey1 (n=yoann@lns-bzn-49f-81-56-185-87.adsl.proxad.net)
21:03.18cr2aeabi ?
21:04.18NetRippertekkdrone, yes the messages are delayed on purpose, so that it's a bit readable and wont go too fast
21:04.28NetRippertekkdrone, i can turn it off though, but for debugging this is better
21:04.46*** join/#htc-linux iNFEKT (n=nFKT@216.163.57.57)
21:05.44*** join/#htc-linux nebi (n=nebi@170.ftth2.cust.fyrobs1.upps.se.borderlight.net)
21:07.37tekkdroneNetRipper: for sure, especially considering boot_delay will skip quite a bit at normal speed
21:10.43tekkdronecr2: were you asking me? I'm not sure, I only glanced at the build options, didn't check to see if they're using arm's aeabi
21:13.44NetRippercr2, for the kernel we use eabi
21:14.12NetRippercr2, i dont know what wince5 compiler does
21:16.49cr2is it known where is the g1 spl located in sram ?
21:17.10cr2has somebody dumped sram on g1 ?
21:19.50excowhat do I need to setup to be able to connect to haret through haretconsole or telnet?
21:22.38*** join/#htc-linux gmarzioz (n=gmarzioz@host60-41-dynamic.61-82-r.retail.telecomitalia.it)
21:23.04excoI always get "Socket error: Connection refused"
21:23.11gmarziozhallo
21:23.34excohi
21:23.45gmarziozwho use blueangel??
21:24.49excoask what you want to know and see if someone responds
21:25.23tekkdronecr2: I don't know, if not and someone can furnish the procedure I have access to a G1
21:25.52tekkdroneexco: I connect the device, run haret, click listen, then connect to the pda's IP via haretconsole.py
21:26.06cr2tekkdrone: can you read /dev/mem on g1 ?
21:26.20tekkdroneI can't even get a root shell on G1
21:26.32gmarziozi have a blueangel, i boot linux from sd, what xserver??
21:26.36cr2hmm. that sucks
21:26.49tekkdroneused to not be the case, t-mob patched it closed
21:26.50gmarziozqtopie is the best?
21:27.02tekkdrone(so google released the dev version of the phone, fully unlocked)
21:27.28tekkdroneunlocked down to the bootloader from what I gather
21:27.31*** join/#htc-linux kimhoon (n=kimhoon@s559116c1.adsl.wanadoo.nl)
21:28.01gmarziozthere is a canal for htc blueangel??
21:28.54gmarziozthe wireless work
21:29.12gmarziozi'dont speak english bien..help
21:30.07excogmarzioz: best - that's just subjective - depends on what you like and what you want to do with it
21:30.33gmarziozexcuse me
21:30.55gmarziozi want work wireless battery charge gps
21:32.05excogmarzioz: http://handhelds.org/moin/moin.cgi/BlueAngel
21:32.28excodon't know if there has been further progress since then
21:32.31cr2hehe
21:32.39cr2dcordes: are you sleeping ?
21:33.06gmarziozi tray that
21:33.20tekkdronegmarzioz: http://wiki.xda-developers.com/index.php?pagename=BlueangelResearch tambien
21:33.55cr2~ping dcordes
21:33.55aptpong dcordes
21:34.50excotekkdrone: I don't have/see haretconsole.py but ./console IP only results in "Socket error: Connection refused" (pinging it works fine)
21:34.53*** join/#htc-linux stefan_schmidt_ (n=stefan@p5B037F1A.dip.t-dialin.net)
21:35.12cr2non-A people around ?
21:35.27jerry507_macWell then you have haretconsole.py
21:35.36tmzttekkdrone: you have a g1?
21:35.48tekkdronemy coworker does, and he's cool as hell
21:36.07tekkdronehe's seen the apache running android so he trusts me to a degree ;)
21:36.14excotelnet 169.254.2.2:9999 telnet: could not resolve 169.254.2.2:9999/telnet: Name or service not known
21:36.14tmzttekkdrone: android-dls.com wiki
21:37.00*** join/#htc-linux Rogro82 (n=rogro82@s5591104d.adsl.wanadoo.nl)
21:37.12tmztexco: telnet ip port
21:37.16tmztexco: no :
21:37.20cr2telnet 169.254.2.2:9999
21:37.29cr2telnet 169.254.2.2 9999
21:37.36Rogro82hiya all
21:37.40excoyeah, meant that ... http://handhelds.org/moin/moin.cgi/BlueAngel
21:37.49excosorry: telnet 169.254.2.2:9999
21:37.49excotelnet: could not resolve 169.254.2.2:9999/telnet: Name or service not known
21:38.02tekkdroneleave out the colon
21:38.03excoforget the link
21:38.04cr2tmzt: i've found the titan A0/A4 formula
21:38.07tekkdrone(if using telnet)
21:38.29tekkdroneif you're using haretconsole, I don't remember if it's necessary to specify the port
21:38.34tekkdroneI don't think it is
21:38.37cr2need to match the bitmasks
21:38.49jerry507_macIt's not
21:38.58jerry507_macI get the same error though
21:39.23*** join/#htc-linux _XD (i=t3st1fy@ircop.com)
21:39.43excotelnet 169.254.2.2 9999
21:39.43excoTrying 169.254.2.2...
21:39.43excotelnet: Unable to connect to remote host: Connection refused
21:39.47jerry507_macI chalked it up to being tired
21:39.53gmarziozhow i can manage wireless
21:39.54_XDwhats up doods
21:40.12excoharetconsole$ ./console 169.254.2.2 9999
21:40.13exco========  Connecting to haret  ========
21:40.13excoSocket error: Connection refused
21:40.50tekkdrone./console 169.254.2.2 gives you the same?
21:41.05excotekkdrone: yes
21:41.16*** join/#htc-linux dcordes_ (n=dcordes@unaffiliated/dcordes)
21:41.17gmarziozexco how i can manage wireless
21:41.39excogmarzioz: sorry, I don't know
21:42.03MullinsRogro82: You have touch screen working on Kaiser Android?
21:42.22cr27x00 -> mask1 -> x,1,a,5,3,3,1,1
21:42.32gmarziozno wireless for htc blueangel
21:42.34gmarzioz???
21:42.57Rogro82I dont have a kaiser so a bit difficult for me :)
21:43.05tekkdroneexco: are you trying this from windows? I've only ever console'd in on the *nix side...
21:43.13excoI just built my first zImage from htc-msm ...
21:43.21excotekkdrone: from Ubuntu
21:43.22jerry507_macI was trying it from windows when I got that error
21:43.29tekkdronetmzt: excellent, I'll go get his phone
21:43.31Rogro82Shouldnt be that hard though i guess
21:43.55cr27x00 -> mask2 -> x,3,64,32,3,3,0,1
21:44.03tekkdroneexco: if that's the case I really have no idea
21:44.06jerry507_macI guess I'll try it from my mac here
21:44.17excofrom windows I managed to telnet when I last tried ... half a year ago on my Magician
21:45.28MullinsRogro82: ooops, thats Anog got that working :)
21:47.08cr27x00 -> mask4,8,10,20 -> x,1,a,5,3,3,1,1
21:47.38tmztexco: is haret still busy?
21:48.03cr27x00 -> mask_other -> x,1,a,5,3,3,1,1
21:48.12tmztcr2: tekkdrone tried the raph100 kernel on raph500 and got as far as switching to console but the usb doesn't work
21:48.15Mullinsdcordes: when you tried kaiser-smd in the vogue tree did it boot up give a few debug messages and power off?
21:48.37tmztcr2: what are those?
21:49.07cr2i need to recheck the divisor settings, but the algorithm looks the same as A
21:49.34cr2tmzt: A0/A4 et al. clk divisor setup
21:49.56excotmzt: haret shows the waiting sign after clicking listen ... then I try to connect but it apparently gets ?refused
21:50.23excoDo I need to set some specific IP because right now it uses dhcp
21:50.57tmztcr2: it appears it wasn't possible to dump g1 spl through /dev/mem, I think /dev/pmem was also tried and failed
21:51.01cr2tmzt: the current raph100 code needs some critical changes, otherwise the gpio setup is fucked up.
21:51.08tmztcr2: it was evantually accessed through mtd
21:51.26cr2tmzt: mtd is not interesting
21:51.41cr2probably they have set some protection bits
21:51.42tmztwhat are you looking for?
21:51.53cr2spl relocation
21:52.11cr2because the mtd code is put somewhere else
21:52.18tmztok
21:52.21tmztdon't know
21:52.57cr2well, the a0/a4 clock setup looks very similar on A and non-A
21:53.07cr2maybe even identical
21:53.13Rogro82Mullins which code in the vogue tree you mean? the new smd or the previous original vogue smd?
21:53.17cr2time to clean this evil mess
21:53.28tmztcr2: would you know how to dump ulpi registers for usb?
21:53.47cr2no
21:54.19tmztwe are trying to find the differences between fsusb and hsusb, also it appears raph500 might have a different hsusb device controller
21:54.40cr2i doubt it
21:54.48cr2it's the same cpu
21:54.48tmztI think maejre*p was the one working on it
21:54.54tmztwhich cpu?
21:55.01tmztmsm7500a?
21:55.08cr27x0xA
21:55.31tmztwe haven't tried the raph800 kernel on it yet
21:55.54tmztit identifies in wm as msm7500a while raph800 identifies in wm as msm7501a
21:56.01tmzts/it/raph500/
21:57.39cr2can you dump the 500 spl ?
21:57.40MullinsRogro82: The one Dzo committed today
21:58.00Rogro82this one doesnt handle smd channel info for the kaiser
21:58.17tmztcr2: don't have one yet
21:58.42tmztthis is from tekkdrone working on it yesterday
21:58.56cr2ok, somebody who has, should do it
21:59.06tmzttekkdrone: can you dump spl with haret?
22:00.05MullinsRogro82: He has a conditional check 'if(machine_is_htckaiser()' in it. He intends it to work for vogue and kaiser AFAIK. vogue-smd.c I am on about
22:00.23tmztDracoDan: are you there?
22:00.36Rogro82i know but there is not kaiser code yet so it only does vogue for now
22:00.54Rogro82s/not/no
22:02.04Rogro82just comment out the machine_is_htcvogue stuff to get it booting ( i guess this will work cause the original vogue code also made kaiser boot ).... modem will not work though
22:02.36Rogro82or use the vogue-smd code from an older commit until DZO releases a new one
22:02.47tmztdo all the msm devices need smd (some) channels manually configured?
22:03.05*** part/#htc-linux piusvelte (n=chatzill@nat.philau.edu)
22:03.05tmzts/smd (some)/(some) smd/
22:03.26MullinsRogro82: Yeah, I got it booting using the Vogue code. just thought that might prevent other stuff working such as keys and touch screen?
22:03.32tekkdronetmzt: probably can shortly, about to root this G1 as soon as I can determine how to backup and/or if this reversioning will hose data on the phone
22:03.56tmztnandroid backup I think
22:04.02tekkdroneI'll look it up
22:04.14tmztnbd erases everything
22:04.21Rogro82Mullins did you compile in the touchscreen driver and the kaiser-keybad board etc?
22:04.21tmztnbh
22:05.10Rogro82s/keybad/keypad
22:05.13Rogro82s/keybad/keypad/
22:05.41MullinsRogro82: tsc2003.c and board-kaiser-keypad.c. Only some keys work and no touch screen at all
22:05.42Rogro82:) hehe apt thats not what i meant :P
22:06.26Rogro82you should check against android-msm to get the right files
22:06.26tekkdronetmzt: thank you, that's critical info since the phone's not mine
22:06.47MullinsRogro82: I can get Dzos commit to start up fine, just cant navigate properly yet
22:08.02MullinsRogro82: BTW, if you don't have a Kaiser how did you get 1.0 to run on it originally??!?
22:08.11tmzttekkdrone: if you want more information, #android is the place to ask (and that wiki), you just don't want to go in there and say "I heard there's a downgrade now" or something like that
22:08.13MullinsRogro82: I am reading YOUR guide!
22:08.35tmztdcordes  has a kaiser
22:08.53Rogro82yould need to mix in some original kaiser code to make things working... check the board-htckaiser
22:09.29Rogro82Mullins: I build it for polaris (almost the same as kaiser)
22:10.26Rogro82Mullins: The guide clear enough? :)
22:10.31tekkdronetmzt: lol, no worries I'm an old school irc'er, etiquette is not lost upon me
22:12.23Rogro82Mullins: At which base are you starting? Are you trying to make the vogue-tree kernel work with Kaiser specific stuff or the Android-msm branch with needed vogue stuff?
22:13.05Rogro82Mullins: Both ways worked for me... Starting at the vogue one was a bit less work though
22:15.12cr2tmzt: hmm. the c0 clkdiv looks like uart clock divisor
22:15.48tmzta0/a4 are sdc1, xx/xx are sdc2, b0/b4 are sdc2?
22:16.07tmztc0 is for uart? is this the bluetooth uart
22:16.35tmzts/sdc2?/sdc3?/
22:18.46cr2tmzt: a8/ac are sdc2, it's all in the wiki
22:19.28cr2tmzt: c0 is probably a prescaler for all 3 uarts
22:19.49MullinsRogro82: I am using the vogue-tree cos Dzo is now attempting to get it to work for Kaiser too. See here -> http://git.linuxtogo.org/?p=groups/mobile-linux/kernel.git;a=commit;h=4726e20e839a1644626560dfa72b3e0e730ba86a
22:19.52cr2but i think it's not really used on htc phones
22:20.39MullinsRogro82: I was working with yours this morning when this commit was put up so I switched to that. It is similar to what you did
22:21.14MullinsRogro82: but I still had to use the vogue board code with some tweaks. I saw that you also had to do that
22:23.15Rogro82Mullins: Yes you had to use smd_sleep_enter/exit etc... the vogue code doesnt work on kaiser but atleast it boots
22:24.21Rogro82Mullins: The merge DZO did is basically the same as i made only i used the polaris code in the vogue-board...
22:25.11j0b0pichurri are you around
22:25.37Rogro82Mullins: The board-htckaiser-old-smd.c code does work but its an old codebase and not stable... I can make a data connection with it but cant get android to boot with it.
22:25.57MullinsRogro82: hmm, vogue-ts.c worked for you on Polaris?
22:26.57Rogro82no vogue-ts is changed with Polaris code. Dwaradzyn created it and only made minor changes but didnt rename it so its still vogue-ts :)
22:27.43cr2NetRipper: ts init at 0xa81* must be done.
22:27.54NetRipperi tried
22:28.02NetRipperbut i dont know init sequence
22:28.15cr2ok, so i'd document it
22:28.37NetRipperif you would please
22:28.37NetRipper:)
22:28.49NetRipperalready found some things..
22:28.54NetRipperbut it is just by guessing
22:29.02NetRipperchanged some values and checked the result;p
22:29.13NetRipperi know how to turn the touchscreen on and off :p
22:29.25NetRipperand 0x10 is some kind of status
22:29.47cr2NetRipper: what about msm irq fe/re/0/1 setup ?
22:29.49NetRipper0x8 and 0x14 offsets cannot be changed.. haret doesn't even reg when i change the memory value
22:30.00NetRippercr2, didn't work.. i tried all, but the irq still doesnt fire
22:30.33NetRippernone of the 3 irqs even.. sbbi, ts1 and ts2
22:30.50NetRipperresp 29, 30, 31
22:30.57DracoDandid someone need some testing done on the RAPH500?
22:31.01cr2i saw ts1 and ts2 in wince
22:31.11NetRipperin only see ts1 in wince
22:31.16cr2DracoDan: dump the spl
22:31.20tekkdronetmzt: I imagine dumping my raph500 SPL won't do much good, I'm unlocked and customized
22:31.35DracoDanim new, youll have to be more specific
22:31.55tmztDracoDan: start haret and connect with putty
22:31.57cr2do you know how to connect to haret ?
22:33.27tmzttekkdrone: who is working on that for raph500?
22:33.34tmzttekkdrone: the spl and bootloader changes
22:33.56cr2NetRipper: the ts init at 0xa81* shuld be done first. then we can talk about the irq
22:34.17NetRippercr2, ok.. already iomapped the memory, so just need to know what to do
22:34.31cr2ok
22:34.38tekkdronetmzt: raphael elite team, there's a SPL restoration package also
22:34.46cr2i see that the wiki is empty
22:34.53tmzttekkdrone: ok, but is it specific to the vzw?
22:34.59tekkdroneyes
22:35.04NetRippercr2, the touchscreen part, yes
22:35.09tekkdroneit unlocks us so we can use the RUU
22:35.18NetRipperyou want me to put my small list of findings on there?
22:35.19tekkdroneand is vzw specific
22:35.33cr2NetRipper: sure
22:36.04cr2NetRipper: now i have checked the SD divisor calculaton on raph/diam/tita/kais
22:36.36NetRipperare they similar?
22:36.41DracoDandamn it, was forgetting to change from ssh to telnet
22:36.43NetRipperthe formula's
22:36.45DracoDanok, ready
22:36.45cr2they all look the same, but we should document all available options
22:36.49pichurrij0b0, hey...
22:36.54NetRipperok
22:37.12cr2pichurri: do you have bstone ?
22:37.21pichurricr2,
22:37.23pichurricr2, yep
22:37.26jerry507_macWhat is the spl anyway?
22:37.43cr2pichurri: can you dump the spl ?
22:37.55DracoDanim connected to my phone and waiting
22:37.58DracoDanraph500
22:37.59jerry507_macFull acronym and I can google it
22:38.07cr2pichurri: are there any hw wiki docs for it ? does the touchscreen work ?
22:38.33cr2DracoDan: pwf r500spl 0x0 0x100000
22:38.50pichurricr2, spl, is the bootloader...right? I' m using hardspl...
22:39.01MullinsRogro82: How do I stop Android from starting so I can use the console for debugging?
22:39.08DracoDannow what?
22:39.38pichurricr2, all I have done with the blackstone is on the wiki, the screen still is unreadable...
22:39.38Rogro82Mullins: Edit the init.rc in Initrd.gz and comment out the zygote service
22:39.41MullinsI've been trying all day to get it to start and now it can , I don't want it to :O
22:39.48j0b0hi pichurri
22:39.54cr2pichurri: it does not matter, dump it with the command above (replace r500spl with bstospl) or something like that
22:40.07j0b0i saw you were looking into the messed up screen thing
22:40.11cr2DracoDan: do you have a 1mb file r500spl in the haret dir ?
22:41.04DracoDanindeed i do
22:41.07j0b0i have been looking into the raph500/diam500 problem with colors
22:41.07pichurrij0b0, hey!
22:41.08j0b0but that is not nearly as severe as what i saw in your picture
22:41.15j0b0that was just coorspace
22:41.19j0b0(and still is :( )
22:41.26cr2j0b0: are they really 12bit ?
22:41.43pichurrij0b0, do you think you can help me to debug the screen output for blackstone...?
22:41.47DracoDando you need the file?
22:41.59cr2DracoDan: yes
22:42.01j0b0well.. when you boot the kernel we currently boot, it goes into a 12 bit mode
22:42.05pichurrij0b0, I have access just via ssh(usbnet), but the screen output goes like http://www.gavero.com/pichurri/blackstone/images/corrupted_output_2008-12-28_4_small.jpg
22:42.25cr2j0b0: hehe
22:42.48j0b0i think the solution would be to somehow put the panel into the correct 565 mode
22:42.54cr2j0b0: and wince is 12bit too ?
22:43.13lupine_85OK
22:43.30j0b0otherwise we'd have to modify pixelflinger, surfaceflinger, libagl
22:43.38j0b0which is doable but not nice
22:43.39lupine_85If we want to know how the Raph100 keyboard works, the example in question is WindowsCE driver KSC1
22:43.58j0b0i dont know. i only have a raph100
22:44.01tekkdronej0b0 would it not be easier just to translate in the fb module?
22:44.02DracoDanhttp://dracodan.dyndns.org:8080/r500spl
22:44.12lupine_85(disabling it makes keyboard input impossible)
22:44.16j0b0i think it is lack of init code
22:44.40pichurrij0b0, I played with htc_fb_console, the update function, the output changes but not in a good way...:P
22:44.59cr2DracoDan: looks good :)
22:45.11DracoDananything else?
22:45.18pichurrij0b0, actually from haret, the output is ok, and I put the fb_base that gives haret, still no good thou...
22:45.23j0b0pichurri in the picture you posted, it was not just colors but totally messed up image
22:45.47lupine_85wonders if different regions are mapped with the keyboard driver unloaded...
22:46.00cr2pichurri: your timing is completely broken.
22:46.40cr2lupine_85: use mmutrace on i2c r/w. the output is very easy to undertsand
22:47.05cr2DracoDan: i think a lot can be done with this dump already
22:47.31j0b0on the raph/diam 500 i could get msm_fb stuff look normal by setting different bit packing values in msm_fb_probe()
22:47.34maejrep[w]cr2: how would I determine if the i2c base is different on raph800?  I mentioned earlier that the phone freezes when I try to trace that area
22:47.38cr2lupine_85: i didn't do it because htere are more critical problems to resolve first :)
22:47.47j0b0the red/green/blue length and offsets
22:48.10DracoDancr2: PM me if theres anything else i can help with, as long as it doesnt put my phone at serious risk
22:48.16DracoDanits my only phone
22:48.42*** join/#htc-linux gmarzioz_ (n=gmarzioz@host60-41-dynamic.61-82-r.retail.telecomitalia.it)
22:48.45j0b0the blitting stuff honors it (the tux logo looks good) and the console honors it (i use those values to setup the palette)
22:49.03j0b0but the software renderer android uses just assumes 565
22:49.36tmztis this on the vic side, mddi, or does the chip connected to mddi need to be initialized somehow?
22:49.43j0b0pichurri but your problem is more severe than bit packing
22:50.08cr2maejrep[w]: i can check the spl. imvho the i2c base is the same
22:50.11pichurrij0b0, do you know why still in haret, if I write to the fb physical address it shows on the output but not on lnx
22:50.28lupine_85well, the keyboard is #1 for me :p
22:50.41lupine_85cr2: was the the memory address and offsets you mentioned yesterday?
22:50.47pichurrij0b0, I think that bit packing is ok, or can't tell, but first thing is to get readable output
22:50.51cr2tmzt: the lcd mddi clock divisor setting is different for most phones, btw
22:51.14maejrep[w]cr2: hmm, so what would explain it freezing the phone? :x
22:51.20cr2lupine_85: look for the virtual remapping of the 0xa99* area
22:51.24cr2lupine_85: the offsets are in wiki
22:51.49NetRippercr2, http://wiki.xda-developers.com/index.php?pagename=RaphaelTouchscreen
22:52.13j0b0yes. i never got any success, but what i was eventually looking at was the dma flags in mdp
22:53.09dcordes_cr2: no. what do you need tested on non-A
22:53.23j0b0pichurri which mddi_client are you using
22:53.30cr2NetRipper: looks good already
22:53.53j0b0in raph/diam its drivers/video/msm/mddi_client_toshiba
22:53.57cr2dcordes_: i have checked the a0/a4 et al calculation on raph/diam/tita/kais
22:54.08pichurricr2, I dumped the slp: http://www.gavero.com/pichurri/blackstone/blac100slp
22:54.34pichurrij0b0, same as the 800, mddi_c_4ca3_0000
22:54.35cr2dcordes_: the formula looks the same for all of them, but we need to document and compare the 7 params
22:54.42tmztcr2: do you know how we can get the wifi eeprom data from nand on wm htc devices?
22:54.42NetRippercr2, however when i boot linux and read those regs, they're exactly the same as in wince, unmodified... i <it crowd style>tried turning it off and on again</it crowd style> (setting 0x0 at +0x18, then 0xa again), but doesn't help
22:55.38pichurricr2, what info can you get from the slp? initialization of the screen? how can I see what is inside?
22:56.05pichurricr2, sorry, spl...
22:56.51j0b0right. i found that arch/arm/mach-msm/board-trout-panel does some init stuff
22:56.54cr2pichurri: looks good too
22:56.59j0b0thats really where i ended up
22:57.08cr2pichurri: the SD clock table is the same as in my raph100
22:57.33cr2uh, interesting
22:57.37dcordes_cr2: ca you be more specific? what you want me to do?
22:57.42cr2i thought 19MHz was mmc
22:57.53cr2SD clk rate 19MHz
22:57.53cr2SD clk rate 144KHZ
22:57.54pichurricr2, how do you know? how did you open it?
22:57.55cr2SD Clk rate 24 MHz
22:57.55cr2MMC clk rate 12 MHz
22:57.55cr2SD Clk rate 48 MHz
22:58.37cr2dcordes_: check the logs, i've added the tita clock masks there (and 7 params)
22:58.39pichurricr2, thats great news for android 1.0, but still no output thou...
22:59.15cr2bstone spl has SREC decoder
22:59.55cr2hm. the same atmegap48
23:00.02cr2Reset Micro_P48(KEY)
23:00.02cr2Reset Micro_P48(LED)
23:00.52cr2qimonda_qhnzmmbzaaazaa22a
23:00.59cr2is it some flash chip ?
23:01.59dcordes_we have the id table from kaiser in wiki
23:02.50cr2maejrep[w]: i don't seen any differences for i2c on raph800
23:02.52dcordes_http://wiki.xda-developers.com/index.php?pagename=MSM_NANDID can't see matches
23:03.08cr2dcordes_: then it's something new
23:03.16cr2EPS_774
23:03.19cr2wtf is that ?
23:04.10cr2some epson controller ?
23:04.10cr2tmzt: what epson datasheet have we found on raph800 ?
23:04.38pichurrij0b0, what controls the timing you were talking about?
23:05.17dcordes_cr2: http://www.qimonda.com/ they produce memory
23:05.18j0b0remind me ..
23:06.03j0b0what timing was i talking about
23:07.21pichurrij0b0, sorry, it was cr2, :P
23:07.28dcordes_ROM-capacity:  512 MiB maybe blackstone has something similar to the movi nand (connected to sdcc) in diamond
23:07.33*** join/#htc-linux Jerry507 (n=jerry507@12-226-245-130.client.mchsi.com)
23:07.52pichurrij0b0, but how can I debug/fix the screen output problem
23:07.54pichurri?
23:08.01pichurrij0b0, any ideas?
23:08.16dcordes_cr2: you have the diamond spl? maybe qimonda appears there, too
23:08.18lupine_85ho de hum, that is a lot of i2c data
23:08.20tmztcr2: you mean the video chip?
23:08.25lupine_85\o/ etc
23:09.27cr2tmzt: yes
23:09.51cr2lupine_85: don't trace the status, only the read and write registers
23:10.09cr2and press/release some button
23:10.21j0b0well .. i dont have any actual knowledge but i did see that fiddling with the flags in mdp_dma_to_mddi can have the effect allmost entirely but not quite unlike the picure you posted
23:10.32j0b0where every other line was in the wrong position
23:10.41cr2lupine_85: don't put the fingers close to the touchpad: it's tracked over i2c too
23:10.55lupine_85mm, b2300008 ?
23:11.29cr2<PROTECTED>
23:11.29cr2<PROTECTED>
23:11.30cr2<PROTECTED>
23:11.31j0b0i think its lack of proper or complete init code in mddi_client_toshiba and it could be interesting to look at what board_trout_panel does
23:11.39pichurrij0b0, yeah, thats how you changed the colorset for raph800, right?
23:11.59cr2j0b0: the raph100 init is 99% documented
23:12.12j0b0i never got raph800 work properly with android
23:12.26j0b0only with msm_fb
23:12.36Rogro82im off to bed... bye all
23:13.12j0b0but that was just colorspace. raph800 already has 'stable' fb output
23:13.28Mullinsnn
23:14.13cr2j0b0: it should be possible to see the 444 and 565 difference by writing into the fb directly
23:14.21pichurrij0b0, do you know why is there a stride? any dma2_cfg value that could do any good...?
23:14.23cr2in wince
23:14.37j0b0im pretty much convinces i could modify android libs to look normal on raph800 but i'd much rather make raph800 process android drawing correctly
23:15.01pichurrij0b0, I've been looking at htc_fb_console, I think its easier to play with that one, what do you think,
23:15.02pichurri?
23:15.16j0b0cr2 yes i did that. X11 honors msm_fb and works when you change the bit packing values
23:15.57j0b0pichurri i think htc_fb_console relies on mddi,mdp working properly
23:16.46j0b0(but please dont mistake me for someone who knows about this stuff .. ive just been fiddling with it)
23:17.03*** join/#htc-linux timebomb (n=tb@e176098059.adsl.alicedsl.de)
23:19.48cr2j0b0: ok, but are you sure that the lcd is capable of 565 ?
23:20.24tmztcr2: I think it was s1d something, I had the pdf somewhere, hold on
23:20.50j0b0when you boot the kernel, first you get the druidu fb console, then you see the handover line, then later at some point just before you see the output in the 'new' console you see a flash of the screen as is was before you softresettted back into winmo
23:21.52j0b0i think that is the point where the lcd panel doesnt get inited
23:21.53cr2http://www.eetasia.com/ART_8800358552_480100_NP_51557314.HTM
23:22.00tmzts1d13774
23:23.04j0b0if the panel canonly do 444 (and also does so in windows) then the only option is to modify surfaceflinger and libagl
23:23.05dcordes_j0b0: in X colors are ok?
23:23.05pichurrij0b0, on the winmo? actually the output from htc_fb_console and msm_fb look the same
23:23.38cr2tmzt: ok, so bstone also does not use toshiba mddi chip
23:23.46dcordes_j0b0: report it in the android mailing list. maybe they want to include something to include the correct color code in their libs
23:23.53j0b0well .. i dont have an 800 but when i change the pixel packing values in msm_fb_probe() that does affect colors in X
23:24.09*** join/#htc-linux Jerry507 (n=jerry507@12-226-245-130.client.mchsi.com)
23:24.29dcordes_s/to include/to detect/
23:24.32*** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz)
23:24.41dcordes_~lart apt
23:24.41aptthrows dcordes_'s poor little doggy off a cliff
23:25.30dcordes_apt: try it and die slowly
23:26.09j0b0the android libs call fb->get_caps() (or something) and explicitly request 565. msm_fb will happily say ok even on a raph800
23:26.25j0b0telling android its 444 is no option.
23:26.28cr2j0b0: the datasheet says 565, 666 and 888
23:26.46cr2yuv 422 and 420
23:27.00j0b0thats for the outside
23:27.42j0b0as i understand, the dma takes 565 and puts it into 565, 666, or 888, tight or loose packed, msb or lsb aligned
23:27.47cr2"royalty free driver source"
23:27.52j0b0but it doesnt take anything but 565 or 888
23:29.11NetRipperj0b0, for that keyboard toggle thing, where you can also turn it "off", i just have to add that when vkeyb_toggle is -1, then also don't show the green/blue square, right?
23:29.15j0b0(the _IBUF_ flags and _DST*_ flags in mdp_dma_to_mddi())
23:29.35j0b0right
23:29.47j0b0i will get you code or diff for that
23:30.09j0b0i just test for that in vkeyb_init and return immediately
23:30.10NetRipperi got the toggle_setup from forum
23:30.15NetRipperah ok
23:30.19NetRipperi'll do that too then
23:30.34NetRipperthat's a better way ;)
23:30.45tmztis part of the init directly of the lcm?
23:31.03pichurrij0b0, do you know why HTC_FB_BASE (which is virtual) is different from wince's?
23:31.53j0b0no. but most virtual addresses are different in win than in linux
23:33.04j0b0i lookup the physical address in msm_iomap.h and p2v() in haret to find virtual address
23:33.12maejrep[w]"when changed to 0x0 while +0x00 is 0x8015, touchscreen is disabled until corrected "  <-- is that how you lock the screen? ;p
23:33.17pichurrij0b0, ok, so thats ok...do you know why there is an offset? if the base for fb physical memory its ok...?
23:33.29lupine_85is going to *spank* a driver out
23:33.34lupine_85...in theory :D
23:33.48pichurrij0b0, I got the physical from haret's VRAM I think
23:33.55dream_killhi guys
23:34.02dream_killany progrss on raph100 :D ?
23:34.17j0b0static int __init msm_vkeyb_init(void)
23:34.17j0b0{ if (vkeyb_toggle == -1) return 0;
23:34.57j0b0pichurri the address seems correct, as yu see stuff on the display
23:35.18j0b0its just messed up, but its there
23:35.50maejrep[w]j0b0: you got my fb dump a few weeks ago, showing that the colors looked correct when interpreted as a 16 bit image?
23:35.54maejrep[w]but looked wrong as 12 bit
23:36.12Mullinsdcordes: I have the Kaiser keypad working in Android, but not the cursor keys for some reason. Any ideas?
23:36.14pichurrij0b0, but you know, with 2 different addresses the output is the same, the raph100(0x16800000) and the blac100(0x16044800)
23:36.17maejrep[w]which tends to imply that its capable of (and does) displaying RGB 565 in windows
23:36.53j0b0ah.. actually.. the fb is 2x as big as the screen
23:37.04dcordes_Mullins: keypad and cursor/front keys are connected in a different way
23:37.12j0b0its 2x as high
23:37.29Mullinsdcordes: Ah! in the board code then
23:37.39dcordes_Mullins: the keyboard uses egpio
23:38.08dcordes_Mullins: did you try SD ?
23:38.28Mullinsdcordes: I don't understand that sorry
23:39.20dcordes_Mullins: look at dzo's commit. he adds htc-egpio. it accesses the extra gpios provided by cpld
23:39.28j0b0at some point i found that if i draw not only on (x,y) but also on (x,y+yres) it doesnt flicker in android (which i therefor assume alternates between those 2 'areas')
23:39.44j0b0i dont think the address is wrong or a problem
23:39.46Mullinsdcordes: cool, I will have a look
23:39.55dcordes_j0b0: the x2 is needed for panning
23:39.56j0b0you see the right stuff, you just see it messed up
23:41.30j0b0dcordes_ technically yes, but it 'pans' not line by line, but 640 lines by 640 lines, affectively alternating between the buffers
23:41.36maejrep[w]cr2: do we know anything about the egpio/cpld for raph?
23:41.57j0b0this is not what i know, its what i guess from my fiddling around with the vkeyb and the msm_fb
23:42.31lupine_85curses - out of caffeine!
23:43.34Spionncurses
23:44.04j0b0pichurri i think the things that could be interesting to look at are the mddi_caps (or mddi_client_caps) the panel returns, the lags in mdp_dma_to_mddi and the init code that the trout panel has and the toshiba doesnt
23:44.35*** join/#htc-linux triplie (n=triplie@78.157.23.246)
23:45.35*** part/#htc-linux dcordes_ (n=dcordes@unaffiliated/dcordes)
23:45.41*** join/#htc-linux dcordes_ (n=dcordes@unaffiliated/dcordes)
23:46.48cr2maejrep[w]: raph does not have cpld
23:47.01lupine_85partitions his SD card, mounting images is getting old
23:48.30j0b0dcordes_: about the 2x height fb and the panning: i noticed that when i put android in landscape mode (by sliding out the keyboard in the raphael :D ) scrolling between desktops is much less smooth. if the double _height_ (not width) fb is for panning you would expect the opposite. not that i have any theory on why it is less smooth in landscape orientation tho...
23:50.53maejrep[w]cr2: ah :)
23:53.00excoI just compiled a kernel using htcpolaris_defconfig and htc-msm branch - but it won't mount my SD card - is that expected?
23:53.15*** join/#htc-linux rmoravcik (n=rmoravci@ip-89-102-255-171.karneval.cz)
23:55.22*** join/#htc-linux Jerry5071 (n=jerry507@12-226-245-130.client.mchsi.com)
23:56.05NetRipperhope im not going to make a habit out of stepping on the usb cable (and launching my raphael that way)
23:56.15NetRipperhappened twice so far
23:56.28maejrep[w]get a shorter cable? :p
23:56.31excoget yourself a wireless usb cable ;-)
23:56.41NetRipperwoot, wireless cable
23:56.46NetRipperthey ought to invent that
23:56.59jerry507_macThey have wireless usb
23:57.18jerry507_macIt's an ieee standard if I recall
23:57.51NetRipperheh ok
23:57.54NetRipperbut no wireless cable
23:58.29jerry507_macAh yes, good point
23:58.42jerry507_macIf you attached a dish to your phone then you could have something close
23:59.04NetRippera shorter cable is probably cheaper

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