Friday, 2019-07-12

T42<adampigg> gm09:20
deathmistmorning :)  btw I got both physical side buttons mapped to back by modifying the kernel synaptics driver, which seems to work flawlessly (for now) https://git.io/fjPyl09:22
T42simondvic was added by: simondvic09:22
r0kk3rzdeathmist: you have obs account?09:27
deathmistr0kk3rz: nope, haven't done anything on OBS / mer yet, just been building everything locally09:28
r0kk3rzdeathmist: ok, you need to talk to lbt and get one09:29
r0kk3rzhttps://build.sailfishos.org/project/show/nemo:devel:hw:oneplus:cheeseburger09:30
r0kk3rzthen i can add you to it09:30
T42<adampigg> r0kk3rz!! good to see you, its not the same since u changes tz!09:30
r0kk3rzi have also been busy lately, but hey hey09:30
T42<simondvic> Hello! Please tell me the solution to the problem during the final assembly of sootfs:09:32
T42<adampigg> we cant always tell you, but we can help09:33
T42<adampigg> given time09:34
T42<simondvic> I can throw off the error log during the final assembly09:34
T42<adampigg> also, search the archives09:34
T42<adampigg> piggz.co.uk/sailfishos-porters-archive09:35
T42<Harsh18262> Is Anyone using server for building?09:35
r0kk3rzwhich server?09:35
spiiroindeathmist: mce evdev mapping is applicable only for keys/switches/etc mce itself is using (i.e. power key, lid/slide switches, to some extent volume keys)09:45
spiiroindeathmist: basically other keys are handled by upper levels, compositor/lipstick, apps, and so on09:46
spiiroindeathmist: power key events coming from touch input devices are assumed to be double tap and handled as EV_MSC:MSC_GESTURE:4 -> which would explain that gesture(4) you see in logging09:47
T42<adampigg> spiiroin, am i right in thinking sfos doesnt use key-bac as a back swipe action09:48
deathmistspiiroin: alrighty, no problem tho as this seems to do the trick perfectly fine for me :p  may look into other solutions later if I ever start doing stuff with android apps on SFOS09:49
spiiroin@adampigg depends... android style home/back/menu keys are handled at ui level - but there are several "*back*" events and only one of those applies09:49
spiiroin(IIRC there was some confusion about which BACK event should be used)09:50
T42<adampigg> spiiroin, on mido, the android keys work in ad, but dont to anything in sfos. event are key_back and key_homepage09:51
spiiroin@adampigg could be that you need to declare them in hw-settings -> hw keys test in csd lists and reacts to them -> ui should too09:52
spiiroin@adampigg /usr/share/csd/settings.d/hw-settings.ini -> [Keys] section, something like09:54
spiiroin[Keys]09:54
spiiroin# Device with volume keys and android style menu/home/back keys09:54
spiiroinKeys=0x01000072, 0x01000070, 0x010000dc, 0x01000090, 0x0100006109:54
spiiroin#    VolumeUp    VolumeDown  MenuKB      HomePage    Back09:54
spiiroinbtw, those are qt side key enumration values from qnamespace.h09:55
deathmistr0kk3rz: I now have a working OBS account under the same name10:02
mal@adampigg your qtbase change just got merged :)10:26
T42<adampigg> spiiroin thx10:26
T42<adampigg> mal, gr810:27
T42<adampigg> need to find a solution to mesa10:27
maldeathmist: I thought I said already yesterdat about that the mapping wouldn't help with back button10:27
mal@adampigg yes, that needs some thinking10:27
T42<adampigg> remember your joblist :)10:27
deathmistmal: well the fix I was thinking of doing in the first place seems to work fine so it's all good :p10:31
deathmistI just wanted to first try a more sane way that wouldn't need kernel level modifications and remapping10:32
r0kk3rzdeathmist: done10:54
deathmistawesome, thanks! now to figure out how all of this works using the FAQ section :)11:01
T42<Harsh18262> @r0kk3rz or gcp or any another I m getting permission denied for fixupmountpoint file it is something related to permissions to the file or something I am using a ext hhd and mounting it in a folder using mount command11:33
malmaybe that is mounted with noexec or something?11:38
T42<Harsh18262> how t check and fix it? right now i am building other roms by using sudo -s but it is not working perfectly for platform sdk14:06
malcheck output of mount command on host to see how it's mounted14:10
deathmistso I started fixing physical button backlights, logcat reveals it's looking for dirs like backlight1 & 2, when there is only "backlight" dir in sysfs and changing brightness there works perfectly, what should I do? https://pastebin.com/yw3QZmXb14:37
T42<Harsh18262> @mal getting this line /bin/bash: hybris/hybris-boot/fixup-mountpoints: Permission denied14:44
mal@Harsh18262 did you do what I said?14:44
maldeathmist: what has logcat to do with backlights?14:44
KALUBEdeathmist: find out how Android does it?14:45
maldeathmist: so you didn't look at https://git.merproject.org/mer-core/mce/blob/master/inifiles/buttonbacklight.ini14:47
T42<Harsh18262> @mal yes14:49
T42<Harsh18262> @mal got this /dev/sdb on /home/harshwardhanmehrotra/hhd type ext4 (rw,relatime,data=ordered)14:49
T42<vince1171> Deathmist: small question, is this your repo?14:51
T42<vince1171> https://github.com/sailfishos-oneplus5/important14:51
deathmistvince1171: yes :p14:51
deathmistmal: that looks very promising, I was looking in the mce repo before but at completely wrong things, I'll bookmark that inifiles dir for future refecence too :)14:51
T42<Harsh18262> @mal fixed for now i did chown -r on hadk folder and chmod 755 on fixupmountpoint file working for now14:51
maldeathmist: a simple search for button backlight would have helped in that repo14:52
deathmistmal: didn't know you could search it too, I'm still learning. soo it does work (after I revert enabled & disabled values), why could the backlight be fully disabled after boot?15:33
deathmisti guess it wasn't enabled in the first place (something to do with light hal and the missing paths in logcat?)15:34
maldeathmist: forget light hal, sfos uses the sys paths directly, it ignores android side15:43
deathmistmal: okay, I'm thinking of making a systemd unit that enables the backlight on bootup since it still is off by default15:45
maldeathmist: which backlight?15:45
deathmistphysical buttons (back & recents)15:46
malwhy would you want it on always15:46
maldeathmist: try using the sysfs path directly from sfos (make sure permissions are suitable)15:46
deathmistmal: doing "echo 1 > /sys/class/leds/button-backlight/brightness" even as nemo works as expected and the button lights go off on lockscreen (except seem to turn on whenever any background task is active and then turn off again)16:01
malok, I haven't used those in a long time16:03
piggzMister_Magister: <cough> keymap </cough>17:40
Mister_Magisterdunno how to make it xd17:40
piggz_lbt: hi, can you patternise https://build.merproject.org/project/show/nemo:devel:hw:fxtec:t5 pls :)21:37
siceloTheKit: is sfos on droid 4 still maintained?21:44
TheKitsicelo, not really. I still have the device, but we won't be able to get 3.0.3 due to glibc update and old Android kernel, while with mainline there is no PowerVR21:45
sicelosucks. so latest is still what's on tmo?21:46
TheKitthere is https://gitlab.com/sailfishos-porters-ci/maserati-ci/-/jobs/100645934/artifacts/browse/sfe-maserati-2.2.1.18-devel-20180923/, but I don't remember if there were particular issues with the build21:49
siceloi might receive a droid 4 in the next couple of weeks, and will be using it for basic mainline stuff, but trying to figure out what OS to run for daily use. seems your sfos port got quite good reviews compared to LineageOS, for example.21:49
sicelothanks for the link. will save it21:50
TheKitunfortunately it's a pretty old device and I could not find good replacement battery. Regarding mainline, PinePhone seems like a much better target by now minus HW keyboard21:55

Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!