**** BEGIN LOGGING AT Thu Dec 19 02:59:59 2013
Dec 19 09:48:40 <ant_work> gm bluelightning
Dec 19 10:00:43 <ant_work> bluelightning: lumag sent long ago a patch for zaurusd adding collie
Dec 19 10:00:46 <ant_work> http://lists.openembedded.org/pipermail/openembedded-core/2011-September/009193.html
Dec 19 10:01:21 <bluelightning> ant_work: really the patch should just be merged upstream
Dec 19 10:01:28 <bluelightning> I can try to arrange that
Dec 19 10:01:40 <ant_work> wait, there is one more thing there
Dec 19 10:02:21 <ant_work> http://cgit.openembedded.org/meta-handheld/tree/recipes-bsp/zaurusd/zaurusd/avoid-rotated-server.patch
Dec 19 10:02:54 <ant_work> imho collie and poodle are different (rotated 270)
Dec 19 10:03:29 <ant_work> (do we use XRANDR in xserver nowadays ?)
Dec 19 10:25:29 <lumag> ant_work, xrandr is just an API which (if I remember correctly) should be implemented by DDX (X driver)
Dec 19 10:28:26 <ant_work> there are so many places in userspace defining rotation...I'm a bit lost
Dec 19 10:28:54 <ant_work> afaik fbconsole is rotated on collie/poodle/akita/spitz
Dec 19 10:28:59 <ant_work> this is kernel
Dec 19 10:30:04 <ant_work> now, the silly calibration in opie is using the mouse interface afais
Dec 19 10:30:15 <ant_work> in kernel we have (collie) CONFIG_INPUT_MOUSEDEV_SCREEN_X=240
Dec 19 10:30:21 <ant_work> so it seems right...
Dec 19 10:31:36 <ant_work> lumag: last reference to kernel are
Dec 19 10:31:36 <ant_work> CONFIG_FRAMEBUFFER_CONSOLE_ROTATION=y and  fbcon=rotate:1
Dec 19 10:31:51 <ant_work> lumag: do you see issues on the kernel side?
Dec 19 10:32:11 <lumag> ant_work, no.
Dec 19 10:32:23 <lumag> I did not yet get to the userspace issues on collie.
Dec 19 10:32:30 <lumag> Still playing with kernel
Dec 19 10:32:41 <ant_work> heh, I'm new to collie userland as well ;)
Dec 19 10:32:54 <ant_work> lacks also alsamixer map
Dec 19 10:33:39 <ant_work> lumag: I have reduced kernel size to 2.4 MB, I'll try to shrink more
Dec 19 10:34:33 <ant_work> lumag: would you keep kgdb and tracing?
Dec 19 10:38:53 <ant_work> bluelightning: anyway, removing the patch forcing calibration even if /etc/pointercal is present, calibration is stuck because of the wrong values and the crosshair is offscreen
Dec 19 10:39:20 <ant_work> so, on collie, the readings are 'wrong'
Dec 19 10:39:36 <ant_work> not surprisingly, collie has a different ts driver
Dec 19 10:40:17 <ant_work> neverthless, on first run the default calibration is precise
Dec 19 10:40:40 <ant_work> I'll just kick out the calibration task on boot
Dec 19 10:46:57 <ant_work> bluelightning: and ehm..no buttons
Dec 19 10:49:52 <bluelightning> ant_work: I can't remember, but I think I added the possibility to add custom buttons through the UI, which you could use to find the correct keycodes
Dec 19 10:50:20 <ant_work> th ekeymap is wrong, no Fn, ....etc
Dec 19 10:50:37 <ant_work> on both poodle and collie so smthg is wrong
Dec 19 15:04:40 <ant_work> bluelightning: do you remember how it starts calibration? Reading ts and opie code I'd say Top-Left is 1st sample...but on my devices it is bottom-left
Dec 19 15:21:22 <bluelightning> ant_work: I don't remember the details I'm afraid no
Dec 19 15:25:38 <ant_work> bbl
Dec 19 22:14:43 <ant_home> bluelightning: opie-calibration failing on c7x0 as well...one hint: top Left is really the first sample (no rotation)
Dec 19 22:15:25 <bluelightning> ah, so rotation isn't taken into account then?
Dec 19 22:15:43 <ant_home> seems related
Dec 19 22:15:51 <bluelightning> hmm, interesting
Dec 19 22:16:02 <ant_home> I'll try spitz at last
Dec 19 22:16:20 <ant_home> I expect it to fail as well, with first sample on bottom left
Dec 19 22:19:12 <ant_home> ..mysteriously working on poodle...
Dec 19 22:23:06 <bluelightning> I'm building a qemu image to see how it behaves there
Dec 19 22:25:12 <ant_home> right
Dec 19 22:26:03 <bluelightning> IIRC it was broken under qemu too
Dec 19 22:26:14 <bluelightning> maybe the rotation is just totally broken even with unrotated device
Dec 19 22:33:17 <ant_home> it's funny that poodle starts calibration on btm left the the rest CCW finally center..accepted
Dec 19 22:46:07 <ant_home> bluelightning: spitz is ok ...??
Dec 19 22:47:32 <ant_home> battery monitor is working
Dec 19 22:47:40 <ant_home> (is not on poodle...)
Dec 19 22:50:47 <bluelightning> interesting, we do now try to read from the standard kernel power interface in sysfs
Dec 19 23:05:24 <ant_home> bluelightning: spitz starts calibration on bottom left as poodle
Dec 19 23:07:38 <ant_home> survived to suspend resume (boot from SD)
Dec 19 23:08:43 <ant_home> sounds is missing
Dec 19 23:10:14 <ant_home> lumag: poor sa1100 has some solw down on kexec
Dec 19 23:10:37 <ant_home> is 10-20 times slower than poodle/c7x0/spitz
Dec 19 23:12:05 <ant_home> both from NOR and CF
Dec 19 23:31:24 <bluelightning> event sounds and some app sounds unfortunately relies upon OSS
Dec 19 23:31:43 <bluelightning> I did port the sound recorder to ALSA
Dec 19 23:32:35 <bluelightning> I didn't feel up to porting the rest of it though; one possibile solution would be to use http://sourceforge.net/projects/osspd/
Dec 19 23:32:50 <bluelightning> assuming we don't just enable OSS emulation in the kernel that is
Dec 19 23:33:20 <bluelightning> that might be a big ask just for Opie though, which is why I figured osspd might be preferable assuming it works
Dec 19 23:36:09 <ant_home> afair oss core is compiled
Dec 19 23:51:03 <ant_home> bluelightning: I think I get it now...
Dec 19 23:51:31 <ant_home> with rotation 270 CW (90 CCW) the top left becomes bottom left
Dec 19 23:51:55 <ant_home> so the rotation is considered when displaying the crosshair but probably not taking the samples
Dec 19 23:52:21 <ant_home> (err...spitz and poodle get it right...)
Dec 19 23:55:31 <ant_home> hm..
Dec 19 23:55:47 <ant_home> fbset on spitz says geometry 480 x 640
Dec 19 23:56:04 <ant_home> on collie 320 x 240
Dec 19 23:59:33 <ant_home> poodle 320 x240 as well (doh..320 240 320 243 16)
Dec 20 00:10:12 <ant_home> sent a patch for kernel size right now
Dec 20 00:10:14 <ant_home> gn
**** ENDING LOGGING AT Fri Dec 20 03:00:00 2013