| buZz | lol, tried that neverball-gles | 02:00 |
|---|---|---|
| buZz | bit buggy :P couldnt even start the game in the end, just ignored all the buttons, music kept playing :) | 02:00 |
| * buZz considering patching the code for a 960x540 resolution option | 02:09 | |
| uvos | huh neverball did stop working | 10:45 |
| uvos | it worked absolutly fine in ddk1.9 | 10:45 |
| freemangordon | uvos: any idea how to dum audion controls from android adb? | 10:48 |
| freemangordon | *dump | 10:48 |
| uvos | audio controls? | 10:49 |
| uvos | no | 10:49 |
| uvos | just dump cpcaps registers? | 10:49 |
| uvos | what are you after? | 10:49 |
| freemangordon | this is allwinner tablet | 10:51 |
| freemangordon | I am missing audio routing | 10:51 |
| freemangordon | like what connects where | 10:51 |
| uvos | so androids hal is called audioflinger, that has plugins | 10:52 |
| uvos | on moto devices essentally everything is implemented in a moto vendor plugin | 10:53 |
| uvos | thats closed source | 10:53 |
| uvos | so theres no real awnser here that generally applicable | 10:53 |
| freemangordon | but i see here audio_policy.conf | 10:53 |
| freemangordon | this is a simple tablet with just a speaker and mic and 3.5 jack | 10:53 |
| uvos | theres probubly a generic alsa plugin, that just dose whatever alsa sais for simple cases, maybe it uses that. still not sure what your after exactly but in this case the information would really just be in dts/board code depending on how old the kernel is | 10:58 |
| freemangordon | 3.x | 10:58 |
| freemangordon | I am after having sound from the tablet :) | 10:58 |
| uvos | well looking at the dts would get you furthest | 11:01 |
| freemangordon | no dts | 11:01 |
| freemangordon | adb in android :) | 11:01 |
| uvos | well looking at proc/asound maybe :P | 11:01 |
| freemangordon | ok, what to look there? | 11:01 |
| uvos | idk i still dont know what you need xD, if this is about adding support for the audio setup to the mainline kernel via snd-soc you need to figure out what chips it uses for the dai and codec | 11:03 |
| uvos | that would be in dts or kernel soruce | 11:03 |
| uvos | proc and sys can give you the names of the chips and dai ports too, but really its not going to be enough | 11:03 |
| freemangordon | I already have that | 11:04 |
| freemangordon | but, how to figure out what is connected to HP jack, for example | 11:04 |
| uvos | for the mapphones we gained this information by reading the kernel source | 11:04 |
| buZz | uvos: well, neverball -kinda- works, but yeah, fullscreen doesnt for instance | 11:33 |
| buZz | oo, i guess i could make a neverballrv | 11:43 |
| buZz | neverballrc* | 11:43 |
| buZz | meh ok yeah, fullscreen 960x540 draws only distorted :( | 11:46 |
| buZz | annnnd, crashes hildon? :( | 11:47 |
| uvos | hmm it just exits with cant create gles surface for me | 12:17 |
| uvos | it used to work quite well, except fullscreen | 12:18 |
| buZz | well, it starts here | 12:18 |
| buZz | i built it new from the git repo | 12:18 |
| uvos | i even wrote a proxy to allow it to use the accelerometer | 12:18 |
| uvos | as an input device | 12:18 |
| buZz | ah, is that in the repo aswell? i just used the master branch, didnt even look for others | 12:19 |
| uvos | sure but i dont remember what i called it xD | 12:19 |
| buZz | ahhh, yeah, a 'beowulf' branch with more commits :) | 12:19 |
| uvos | sec | 12:19 |
| buZz | found it ;) | 12:19 |
| buZz | https://github.com/maemo-leste-extras/neverball-gles/tree/maemo/beowulf | 12:19 |
| uvos | https://github.com/maemo-leste-extras/iio-uinput | 12:20 |
| uvos | ah you wanted neverball itself not the prox | 12:20 |
| uvos | y | 12:20 |
| uvos | yeah thats there too | 12:20 |
| buZz | the new branch doesnt seem to have many changes? just some versionnumber that changed | 12:20 |
| uvos | right theres an issue with how it generates the version number is all | 12:21 |
| uvos | that incomapabil with how debian buils the package | 12:21 |
| uvos | there are no other changes in that repo | 12:21 |
| uvos | besides building the gles version instead of gl | 12:21 |
| buZz | uvos: hmhm, well, the master branch built fine for me :) | 12:29 |
| buZz | just did> debuild -us -uc | 12:30 |
| buZz | heh, what a cute thing ; https://www.crowdsupply.com/sutajio-kosagi/precursor | 12:38 |
| buZz | way too expensive imho, though | 12:38 |
| uvos | buZz: the problem is on ci | 13:06 |
| buZz | yeah i have no clue how the ci works really :P | 13:15 |
| uvos | Wizzup: console=/dev/null rw mem=1024M@0x80000000 vram=10300K omapfb.vram=0:8256K,1:4K,2:2040K init=/init ip=off mmcparts=mmcblk1:p7(pds),p15(boot),p16(recovery),p17(cdrom),p18(misc),p19(cid),p20(kpanic),p21(system),p22(cache),p23(preinstall),p24(webtop),p25(userdata) | 14:23 |
| uvos | thats edison | 14:23 |
| uvos | no utags | 14:23 |
| uvos | same story in cdt.bin | 14:23 |
| uvos | looks like you cant download edison fw images anymore anywhere i can find | 14:34 |
| uvos | but i archived them before the dissapeared | 14:34 |
| uvos | at least some of them | 14:34 |
| uvos | i have 2 different version of allow-mbmloader-flashing-mbm | 14:34 |
| uvos | http://uvos.xyz/maserati/mbm/edison/ | 14:34 |
| uvos | i also have dts for the global variant (me865) http://uvos.xyz/maserati/stockinfo/dts/me865.dts | 14:35 |
| uvos | looks like it uses the same android version as bionic | 14:36 |
| uvos | its very possible it will just boot a bionic rom | 14:36 |
| Wizzup | uvos: thanks, I feel a bit sick today, so it might be 1-2 days until I get to it | 15:51 |
| uvos | Wizzup: gute besserung! | 16:31 |
| freemangordon | uvos: it seems what I need is a tool called 'tinymix', but I cannot find it pre-compiled | 16:34 |
| uvos | so compile it | 16:34 |
| uvos | geting the android toolchain is tivial | 16:34 |
| freemangordon | yeah | 16:35 |
| freemangordon | this is the only toolchain I don't have around :) | 16:35 |
| uvos | https://github.com/tmlind/cpcaprw/pull/1/files | 16:36 |
| uvos | see readme changes | 16:36 |
| uvos | depending on the age of your android you might have to download an older version | 16:38 |
| uvos | they drop support for old versions of bionic eventually | 16:38 |
| freemangordon | 4.4 | 16:38 |
| freemangordon | but, I will try to find some binary first | 16:38 |
| uvos | android 4.4 is api level 19 or 20 depending on subversion | 16:39 |
| uvos | so if the modern ndk refuses to create a toolchain for api level 19 use an older one | 16:40 |
| freemangordon | I have found this https://apkcombo.com/alsamixer-root/com.skvalex.alsamixer/ | 16:40 |
| freemangordon | uvos: that works,but lists about 30 controls, so I am even more confused :D | 17:03 |
| freemangordon | oh, sorry, not 30, 58 | 17:04 |
| sicelo | fun fact - droid4 accelerometer can work with lis3lv02d (which N900 is currently using), and thus the games can have joystick right away | 18:01 |
| sicelo | 58 controls is child's play (compared with controls on N900, that is) | 18:03 |
| buZz | nice! | 18:05 |
| freemangordon | sicelo: could be, if you know what it is about :) | 18:19 |
| uvos | sicelo: sure but the kernel exposing accel events on evdev is depricated so having a userspace utility do the same thing where needed only is mutch better | 19:57 |
| buZz | should be lower powerdraw anyway | 20:09 |
| uvos | potenutally | 21:25 |
| uvos | as the iio interface supports polling rate changes via standart interface and allows hw triggers to be used | 21:25 |
| uvos | we use both but on d4 neither are available due to the driver being not ideal. | 21:25 |
| uvos | bionic dose both iirc | 21:25 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!