**** BEGIN LOGGING AT Wed Jun 15 02:59:58 2011 Jun 15 09:22:26 bluelightning: do you have some knowledge wrt kernel keymaps? Jun 15 09:23:00 ant_work: not a lot I'm afraid Jun 15 09:23:37 np, I'll try to ask on zaurus-devel Jun 15 13:20:34 Jay7: somehow I can enter System Menu on poodle... Jun 15 13:21:08 the secret isjust pressing the ok key and not the button in the middle of the cursors Jun 15 13:21:31 that sucker key gives out 3 different values (showkey) Jun 15 13:21:44 probably the event interface is confused by that Jun 15 13:22:32 I suspect this is a bug wrt locomo Jun 15 13:22:41 so collie should exhibiy it too Jun 15 14:04:57 the central button does: 0 press, 1 release, 57 release, 0 release, 1 release, 57 release Jun 15 14:08:26 well, not on every reboot :/ Jun 15 14:09:17 the constant seems to be having only one image on mtd Jun 15 14:09:27 no CF, no SD at all Jun 15 14:09:48 then it doesn't freeze Jun 15 14:10:11 so it's probably menu count Jun 15 14:13:41 ant_work: I've seen that before on other devices with recent kernels Jun 15 14:14:04 it seems wrong to me... I don't think kernel keymaps can translate that into a single keycode Jun 15 14:14:15 well, at least a user-space keymap can't Jun 15 14:14:31 my bad I hacked keymap to have the central key acting like OK Jun 15 14:14:43 in a wrong way :/ Jun 15 14:14:47 how did you do that though? Jun 15 14:15:03 I tried to copy from 2.6.24 keymaps Jun 15 14:15:33 I somehow hate this kb stuff :/ Jun 15 14:17:06 imagine on locomo-keyboard devices (collie and poodle) I lack numbers and pipe and minus keys... Jun 15 14:17:31 hm :/ Jun 15 14:17:44 in kexecboot *and* in userspace Jun 15 14:18:14 there must be some obvious kernel change with 2.6.3x wrt keymaps Jun 15 14:18:48 add to that the new keyboard matrix-driver Jun 15 14:18:58 (for clamshells) Jun 15 14:19:26 result: dead keypresses Jun 15 14:51:17 maybe this Jun 15 14:51:35 It seems that in the 2.6.37 kernel the default keymap has moved Jun 15 14:51:37 from drivers/char/defkeymap.c to drivers/tty/vt/defkeymap.c. Jun 15 20:31:43 ping Jay7 Jun 15 20:49:20 anyone have any idea why this might be occuring: Jun 15 20:50:01 kexec -l --append="blarg..." /boot/zImage Jun 15 20:50:28 reboot ... "done, booting kernel." death. wordless, soundless death. Jun 15 22:13:00 Jay7: almost finished klibc tests Jun 15 22:13:11 have you some debug code ? Jun 15 22:13:27 I'd test on corgi otherwise Jun 15 22:13:39 well.. I can write patch quickly Jun 15 22:13:49 (still not tested number keys) Jun 15 22:13:56 no numbers on locomo :p Jun 15 22:14:18 btw lumag is working on locomo Jun 15 22:14:33 ping him about collie+poodle, just in case ;) Jun 15 22:32:37 * Jay7 -> sleep Jun 15 22:32:44 will create patch tomorrow Jun 15 22:32:49 too late and too tired Jun 15 22:42:41 gn Jun 15 22:52:56 btw I have a collie now in addition to an akita and my original spitz Jun 15 22:53:16 (the akita belongs to work though) Jun 15 22:54:06 the akita ought to work ;) Jun 15 22:54:16 insist with the poor collie pls Jun 15 22:54:30 insist? Jun 15 22:54:58 power it of now and then ;) Jun 15 22:56:21 I'll test today's opie in a couple of mins, flashing now Jun 15 22:58:17 booting Jun 15 22:58:38 wow, first udev run is taking ages Jun 15 22:58:56 smthg wrong :/ Jun 15 22:59:12 can't rename /dev/shm/uname Jun 15 22:59:55 calibrating opie... Jun 15 23:00:14 still same issue..bad values are passed and calibration is looping Jun 15 23:00:34 s@i7 Jun 15 23:01:03 well, at least there are crosshairs and ts responds ;) Jun 15 23:02:09 Sysrq, I love you Jun 15 23:10:59 hmm :( you know I think there may be something wrong lower down with touchscreens Jun 15 23:11:14 I had the same problem on rokre6 Jun 15 23:11:25 heh, even better than x11-gpe calibration: no crosshairs Jun 15 23:11:43 the proof would be ts_test I guess Jun 15 23:11:55 and wrong pointercal file at least for c7x0 Jun 15 23:13:16 xinput_calibrator Jun 15 23:13:29 Error: no calibratable devices found. Jun 15 23:14:29 xtscal: xcb_io.c:515: _XReply: Assertion '!dpi->xcb->reply_data' failed Jun 15 23:14:41 Aborted Jun 15 23:15:04 ts_* is not installed anymore Jun 15 23:15:39 so both xinput_calibrator and xtscal are failing Jun 15 23:16:07 argh Jun 15 23:16:39 ADS7846 Touchscreen: no calibration data available Jun 15 23:17:08 this was xinput_calibrator_get_hal_calibration Jun 15 23:19:11 hal ? Jun 15 23:19:42 pointercals are a waste of time IMHO... I spoke to the guy who comitted them to poky, he said he added them to save time during development and "nobody complained they caused issues" Jun 15 23:20:24 dunno what happened since then but I've always found them to be wrong which is why there is a patch in OE to force calibration on first start even if pointercal exists Jun 15 23:20:33 unfortunately x is misplaced about 80 pixel too right Jun 15 23:20:34 (in opie) Jun 15 23:21:04 iirc hrw committed that long ago Jun 15 23:21:38 anyway, really too late now Jun 15 23:21:50 I have to sleep a bit Jun 15 23:21:52 gn Jun 15 23:22:37 ant__: night Jun 16 01:27:58 evenin Jun 16 01:30:51 works with mips/mipsel? Jun 16 01:31:21 haven't been able to confirm patches for kexec-tools working with those archs **** ENDING LOGGING AT Thu Jun 16 02:59:57 2011