Friday, 2020-06-05

T42<edp_17> @elros: Does it matter which bluez I use 4 or 5? What's the difference and the benefit using one of them over the other. Which one is suggested? Thank you.00:24
T42<elros34> @edp_17 I think you should use what everybody else use so bluez 5 unless you have old kernel (I think <=3.4). Also bluetooth-rfkill-event instead these udev hacks I see in your repo. I told you that you should not run time consuming scripts in udev rules.01:21
T42<elros34> <=3.4)01:22
T42<elros34> Oh and remove these udev rulles and use bluetooth-rfkill-event01:23
T42<edp_17> Well, currently I am using bluez4 because I thought with that I could make the bluetooth work. I was wrong, so I go back and try with bluez5. In the meantime I got totally lost in rild, ofono, ril_subscription and related stuff but couldn't work out why rild is crashing. 😔 Without help I don't have the knowledge to solve this.01:26
T42<edp_17> Using bluez5 will make any difference regarding rild and ofono?01:26
T42<edp_17> Thank you for your response by the way!01:27
T42<elros34> no, I cant see conection between bluetooth and radio01:28
T42<edp_17> That is good news. Thanks.01:28
T42<elros34> btw don't add tmp, installroot, documentation.list. They are dynamic and created when you build package01:31
T42<edp_17> Somehow I was able to get an output from '/usr/lib/ofono/test/list-modems' but I can't remember what I changed in ril_subscription and since the command doesn't give me anything.01:32
T42<edp_17> Okay. Now it is too late for me, I barely can see. So, thank you and good night.01:34
T42<elros34> see ya01:34
*** kimmoli_ is now known as kimmoli04:32
riniguspiggz: morning! Can you please tag also test app?05:37
rinigusMaybe later we can have some more user oriented app name and polish bit ui. As it will probably be used as main fp entry point on my port.05:40
riniguspiggz: do you have fp test app icon as svg? So we could use it for cover of the app05:41
T42<Verevka86> (Photo, 720x1280) https://irc.thaodan.de/.imgstore/BjkF1BL2Jw.png06:02
T42<Verevka86> @adampigg @rinigus Big thanks 🤝06:02
rinigus@Verevka86: enjoy!06:20
riniguspiggz: I will make few smaller changes in test app06:23
T42<adampigg> Rinigus, will add svg to repo06:48
rinigus@adampigg: I'll probably submit PR soon, no need to release before that06:48
T42<adampigg> Sure... No need to release before waking up either06:49
piggzrinigus: test app looks good08:13
piggzrinigus: you still need the svg?08:14
riniguspiggz: not necessary, but maybe good to have in the sources. if you have it08:15
piggzi do....08:15
riniguspiggz: thanks!08:16
piggzrinigus: here you go sailfish-fpd-community-test.svg08:19
piggzperhaps we should p.s.a?08:19
riniguspiggz: yes, now all tagged, should be ready for p.s.a08:20
rinigusplease do08:21
piggzPSA: To all community porters.  If you want to bring fingerprint support to your device, please try https://github.com/piggz/sailfish-fpd-community.  See myself or rinigus for help/advice.  Requires android 8+ base.  PRs welcome, this a community project.08:24
deathmistfor OBS I assume you just drop droid-biometry-fp RPM in your droid-hal packaging and call it a day? :p09:32
deathmistalso the init rc file (https://github.com/piggz/sailfish-fpd-community/pull/12#issuecomment-636517230) from yesterday shouldn't be needed anymore, right?09:36
piggzdeathmist: correct, the file is not needed now09:41
piggzdeathmist: and yes for the rpm, just like the other local build ones09:42
piggzjusa: hi, i have an mtk device, running audiosystem-passthrough, audiosystem-passthrough-dummy-af, pulseaudio-modules-droid-hidl .... on calls, i can hear other side, but other side cant hear me, any suggestions?10:10
deathmistpiggz: everything seems to work perfectly so far :) I'll make a new CI image and test that as well before making an announcement on the port XDA page, thanks for your work on this!11:32
T42<edp_17> Hi all, I have a question and a problem.11:34
T42<edp_17> I have switched to bluez5 as I was advised.11:35
T42<edp_17> But when I tried to install the earlier generated broadcom-bluetooth rpm, I got a message "requires bluez-libs" which is not installed indeed.11:36
T42<edp_17> Then I thought I needed to rebuild broadcom-bluetooth and bluetooth-rfkill-event packages so I tried in Platform.11:36
T42<edp_17> But when I ran "rpm/dhd/helpers/build_packages.sh --mw=https://github.com/mer-hybris/broadcom-bluetooth -s rpm/broadcom-bluetooth.spec", I got "bluez-libs-devel is needed by broadcom-bluetooth-1.0.4-1.armv7hl"11:37
T42<玉宸 %lastname%> Fwd from Gavin %lastname%: saw this one is quite interesting, just watch and can get free bitcoin, anyone interested,can you use my referral code, hhh;https://www.okex.com/academy/en/earn-free-bitcoin-while-staying-home?channelFlag=ACECK250147711:38
T42<edp_17> Then I tried to install the "bluez-libs-devel" but got "Problem: bluez-libs-devel-4.101+git77-1.4.1.jolla.armv7hl requires bluez-libs = 4.101+git77, but this requirement cannot be provided not installable providers: bluez-libs-4.101+git77-1.4.1.jolla.armv7hl[jolla]" and a 3 option to remove some packages or break the current installati11:40
T42on or do not install at all.11:40
T42<edp_17> What am I missing? I thought for bluez5, I still need the broadcom-bluetooth and bluetooth-rfkill-event packages.11:41
deathmistanyone else experienced the call UI getting stuck? I just made a call and couldn't close it until the other end ended the call. I was nowhere near the proximity sensor and pressing the power button just kept revealing the UI and it blanked very quickly every time. this is SFOS 3.312:04
ggabrieldeathmist: very rarely, and often while running some sort of heavy process like tracker12:28
deathmistoh that could actually be it, I just rebooted the phone before making the call and I also just reflashed it before that to test other things12:31
rinigusDeathmist: you also need to have sailfish-fpd-community among your packages at OBS. -test could be also of interest12:36
rinigus... in addition to droid-biometry-fp12:36
deathmistrinigus: done that, they're all packaged and I'm about to test the new CI image :)12:37
deathmistexisting users can upgrade with zypper ref && zypper dup and choosing option 1 to deinstall the conflicting encsfa packages12:37
deathmisthttps://github.com/sailfishos-oneplus5/droid-config-cheeseburger/commit/96ddb890bbbf86438d801b4da18c0ef7ad562d9a and obviously I added everything to meta-package / patterns12:38
rinigusI am testing now if we can get this uninstall automatically. let's see if it's possible12:39
deathmisthttps://i.imgur.com/ju6tpsx.png it came up with the FP dialog on initial setup and everything worked just fine ^^12:43
spiiroindeathmist: " power button just kept revealing the UI and it blanked very quickly" = what p-sensor stuck to "covered" would cause - try triple-power-key-press12:43
deathmistah, I'll have to remember that in the future if this happens again12:44
T42<Verevka86> @deathmist [anyone else experienced the call UI getting st …], Work good Xiaomi Mi612:55
T42<elros34> @edp_17 use bluez5 spec files (check PR) for both packages13:00
piggzdeathmist: yeah, works on initial boot on volla too13:29
T42<ankaos> my device is camera not found13:35
T42<ankaos> how fix this problem?13:36
piggzspiiroin: could I map KEY_F11 to turn on screen?13:59
piggzrinigus: my FP also gives out key events when tap/swipt14:00
piggzs/swipe14:00
piggzthers that syntax to map keys, lemme see14:00
piggzspiiroin: reading what was suggested to rinigus, my FP gives a bunch of KEY_FXX events, not gesture events... what is the best course of action for wakeup when touching the FP sensor?14:21
KabouikDo we have an equivalent of libdbus-1-dev under another name?14:26
KabouikNot sure if libdbusaccess or libdbus-qeventloop-qt5 are the same thing14:26
spiiroinpiggz: mapping to power key should work, but due to gemini pda ... KEY_ESC is sort of semi-power key too - you could try if one is better than other15:23
spiiroinpiggz: the "semi" part = always wakes up screen, but does not blank unless mce is pretty sure UI is not using it for some other purposes15:25
riniguspiggz: it seems like double tap could be useful. not always detected though for me, but I am usually triggering pickup gesture when I pickup the phone (not related to fp)15:28
piggzrinigus: spiiroin: if i keep the finger on the sensor for a second, i fet key_f1215:29
piggzs/key15:29
piggzspiiroin: so, i have this file but it doesnt work15:29
piggzsh-3.2# cat 99-fp-power.ini15:29
piggz[EVDEV]15:29
piggzKEY_F12=KEY_POWER15:29
rinigusthat's probably hold15:29
piggzspiiroin: evdev says this about the device https://paste.mozilla.org/oxMUqrqe15:31
piggzrinigus: its easier to trigger that, than the others15:31
piggza tap gives F11, but sometimes you get a swip, whcih is f1-415:32
spiiroinpiggz: and you restarted mce after installing the ini-file?15:33
piggzspiiroin: yes15:33
spiiroinhave you checked that you get down and up events?15:33
piggzspiiroin: https://paste.mozilla.org/J9Y60wc215:34
spiiroinpiggz: looks ok15:36
spiiroinyou could try something like: mce -Tq -levent-input.c:evin_iomon_keypress_cb -lpowerkey.c:*15:36
piggzdid i get the syntax right15:36
spiirointo see if something sane happens15:36
piggznope, nothing there when i hold finger15:37
piggzhow can i check if mce reads and process the config file?15:39
piggzand if mce even ready from event3?15:39
spiiroinevdev_trace -I -> shows what processes have files open15:40
spiiroinmce -Tq -levent-input.c:evin_iomon_device_add --auto-exit=015:40
spiiroinshows how mce classifies the input device15:40
piggzspiiroin: mce: T+0.000 N: event-input.c: evin_iomon_device_add(): /dev/input/event3: name='ff_key' type=USER ACTIVITY ONLY15:40
spiiroindoh15:41
piggzthatll do it :)15:41
spiiroin... and obvious stuff. the ini is in /etc/mce dir?15:42
piggzsh-3.2# cat /etc/mce/99-fp-power.ini15:42
piggz[EVDEV]15:42
piggzKEY_F12=KEY_POWER15:42
spiiroinlooks ok, and mapping should be used in probing already15:42
spiiroinI guess it slips through gaps15:46
spiiroinodd... need to stare code a bit15:47
piggzspiiroin: thanks v much15:47
*** redj_ is now known as redj17:33
piggzspiiroin: am I to assume that the device type is determined before the mappings are loaded which convert key_f12 to key_power18:48

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