T42 | <Bard %lastname%> (Photo, 758x574) https://irc.thaodan.de/.imgstore/569ZHGoiXS.png TG群发,私聊,在线,增粉;三万优质中文群全网监听服务限时特价;需要请咨询; | 06:08 |
---|---|---|
Umeaboy | Even though I'm running LOS-17.1 and the SailfishOS zip is built based on Android, is there a workaround to get Sailfish installed anyway? | 09:24 |
Umeaboy | Lineage is based of Android, but they have to be different in some way. | 09:24 |
piggz | jusa: around? | 09:58 |
T42 | <Anton %lastname%> Hello everyone. I've made changes to some files in device/VENDOR/DEVICE/. What should I do to include them in the build? | 12:41 |
mal | Anton rebuild hybris-hal in HABUILD_SDK and then in platform sdk rebuild droid-hal package using "build_packages.sh -d" | 13:34 |
T42 | <Anton %lastname%> I've found out that devices fails to suspend because wcnss_wlan, apparently. I've tried sailfish-connman-plugin-suspend to no avail. Here're some logs if anyone's interested: https://pastebin.com/cA64EVAr | 17:07 |
T42 | <Anton %lastname%> Oh, and if the solution is about proximity sensor, it's broken as well, I think (calls don't dim the screen, other than that don't know how to test) | 17:10 |
T42 | <elros34> you can test whether proximity works at all in csd. If it fails only in calls then there is some workaround in mce | 17:15 |
T42 | <Anton %lastname%> Prox sensor works, not in calls though. It doesn't bother me now as the sleep issue I was talking about just above | 17:20 |
T42 | <elros34> this might help you for proximity issue: https://github.com/mer-hybris/droid-config-sony-nile/blob/master/sparse/etc/mce/60-proximity-sensor-nile.conf | 17:23 |
T42 | <Anton %lastname%> @elros34 it seems to be no effect, and according to commit comment should pose no effect in my situation | 17:31 |
T42 | <elros34> AFAIK that should help if proximity works in csd/messwerk but fails to work properly when you call to somebody | 17:36 |
T42 | <Anton %lastname%> Could that somehow be connected to ngfd? | 17:42 |
T42 | <elros34> not really | 17:43 |
T42 | <Anton %lastname%> mce seems to be logging "mp_datapipe_set_proximity_sensor_actual(): state: OPEN -> UNDEF" to status when display is powered on. Should I be concerned? | 17:52 |
T42 | <Anton %lastname%> nvrm, it logs that no matter the device state | 17:56 |
T42 | <elros34> I think that is expected with on_demand mode. Proximity is no needed when display is powered on. | 17:56 |
T42 | <elros34> unless you are calling | 17:57 |
T42 | <Anton %lastname%> It constatly switches between OPEN and UNDEF when screen is off | 17:57 |
T42 | <elros34> Does it ever report CLOSED by mce? | 17:59 |
T42 | <Anton %lastname%> Nope | 17:59 |
T42 | <Anton %lastname%> Even if I push the power button with it closed | 18:00 |
T42 | <elros34> Is that moto? Check what values evdev_trace -t reports when you cover/uncover proximity sensor | 18:01 |
T42 | <Anton %lastname%> Looks like silence to me | 18:05 |
T42 | <Anton %lastname%> And yes, that's moto g7 | 18:05 |
T42 | <elros34> does evdev_trace without arguments shows any proximity device? | 18:07 |
T42 | <elros34> I have ct406 proximity driver. If you have some then driver reports 30/1000 instead 0/1000 so that might confuse sensorfw (maybe also mce I dont remember anymore) | 18:08 |
T42 | <Anton %lastname%> evdev_trace reports no proximity sensors that I could recognize. There are a couple Moto CapSense's and a couple duplicate buttons, but nothing remotely proximity | 18:10 |
T42 | <elros34> so maybe this is iio device but I guess mce use it via libhybris anyway. So probably the best would be to run mce in some more verbose mode during call | 18:29 |
T42 | <Anton %lastname%> What should I be looking for there? Running mce with lots of 'v's just prints out lots of variables | 18:34 |
T42 | <Anton %lastname%> Well, it has found those CapSenses as proximity, but they don't change their state with sensor being closed or opened, even if i'm testing in csd | 18:42 |
T42 | <elros34> try something like this to reduce logging to only relevant modules: mce -Tq -l mce-sensorfw.c:*_ps* -l modules/proximity.c:* | 18:49 |
T42 | <Anton %lastname%> It sees something happening, but that's what it got: https://pastebin.com/L4n8t91T | 18:55 |
T42 | CCC %lastname% was added by: CCC %lastname% | 19:10 |
T42 | <elros34> you only pasted few lines so Im not sure but looks like event from sensorfwd are ignored. Maybe because those false "CapSenses as proximity" detection you mentioned | 19:22 |
T42 | <Anton %lastname%> Here's some verbose initialization I was talking about: https://pastebin.com/bebe59KA. I might have to block it from reaching into events1-3, if I've got you right | 19:27 |
T42 | <elros34> try to blacklist those Moto CapSense Ch* devices: https://irclogs.sailfishos.org/logs/%23sailfishos-porters/%23sailfishos-porters.2018-03-22.log.html#t2018-03-22T12:51:06 | 19:34 |
T42 | <Anton %lastname%> Looks to have helped, going to add that to sparse | 19:40 |
T42 | <Anton %lastname%> wcnss_wlan still fails to suspend. Dmesg is flooded with that: https://pastebin.com/X5S7BNWm | 20:00 |
T42 | <erfanoabdi> rinigus: Hey i saw your commit on sensorfw about compass | 21:09 |
T42 | <erfanoabdi> i want to use Orientation sensor to get azimuth for compass | 21:09 |
T42 | <erfanoabdi> https://github.com/erfanoabdi/sensorfw/blob/bullseye/adaptors/hybrisorientationadaptor/hybrisorientationadaptor.cpp#L87 | 21:09 |
T42 | <erfanoabdi> HybrisOrientationAdaptor is sending sensor data in buffer as "CompassData" | 21:09 |
T42 | <erfanoabdi> but my problem is HybrisOrientationAdaptor isn't compasssensor it's registered as Orientation sensor, and should have "PoseData" in buffer https://github.com/erfanoabdi/sensorfw/blob/bullseye/sensors/orientationsensor/orientationsensor.cpp#L114 | 21:09 |
T42 | <erfanoabdi> this makes OrientationAdaptor completely useless | 21:09 |
T42 | <erfanoabdi> `Error occured while reading data from socket: "Unknown error"` when using it as orientation sensor | 21:09 |
T42 | <erfanoabdi> and | 21:09 |
T42 | <erfanoabdi> `main: Failed to create SensorfwCompassSensor: Could not create sensorfw backend` | 21:09 |
T42 | <erfanoabdi> when using it as compass | 21:09 |
mal | @erfanoabdi downgrade the sensorfw plugin (binder or hybris which ever you use) to 0.11.9 | 21:33 |
mal | @erfanoabdi or is that some other issue | 21:34 |
T42 | <erfanoabdi> i'm on 0.11.8 | 21:34 |
mal | ok | 21:34 |
mal | so some other issue then | 21:34 |
mal | what is the issue exactly | 21:34 |
T42 | <erfanoabdi> my question is i'm not sure how to use HybrisOrientationAdaptor | 21:35 |
T42 | <erfanoabdi> it's Orientation sensor but doesn't parse buffer with PoseData, instead it's using CompassData | 21:36 |
T42 | <erfanoabdi> i'm confused what to use in my program when i call dbus | 21:36 |
mal | which device? | 21:41 |
mal | ehat is PoseData? where is the defined? does that even contain correct information | 21:42 |
T42 | <erfanoabdi> this is PoseData https://github.com/erfanoabdi/sensorfw/blob/439062c46182a5955fcb68c8b4e3e673f5539c84/datatypes/posedata.h#L34 | 21:46 |
T42 | <erfanoabdi> class Device orientation. | 21:46 |
T42 | <erfanoabdi> but HybrisOrientationAdaptor is using CompassData, https://github.com/erfanoabdi/sensorfw/blob/439062c46182a5955fcb68c8b4e3e673f5539c84/adaptors/hybrisorientationadaptor/hybrisorientationadaptor.cpp#L83 | 21:46 |
T42 | <erfanoabdi> CompassData only used on compasssensor and not orientationadapter | 21:46 |
T42 | <erfanoabdi> i'm not sure how sailfish handles compass maybe with accel + mag solution | 21:47 |
T42 | <erfanoabdi> but in android we can directly use azimuth of orientation | 21:47 |
mal | I think the virtual compass sensor is broken in sensorfw | 21:48 |
mal | one way would be the grab the compass PR to qtsensors and try that https://git.sailfishos.org/mer-core/qtsensors/merge_requests/8 | 21:48 |
T42 | <erfanoabdi> ah thats using accel + mag | 21:50 |
T42 | <erfanoabdi> we can directly use orientation, but i don't understand how sensorfw registers it.. | 21:50 |
mal | https://github.com/mer-hybris/droid-hal-configs/blob/master/sparse/etc/xdg/QtProject/Sensors.conf | 21:52 |
mal | you replace sensorfw.compass with generic.compass | 21:54 |
T42 | <Malaysia %lastname%> 𝑾𝒉𝒊𝒍𝒆 𝒚𝒐𝒖 𝒅𝒐𝒖𝒃𝒕 𝑰 𝒂𝒎 𝒎𝒂𝒌𝒊𝒏𝒈 𝒎𝒐𝒏𝒆𝒚, 𝑰 𝒔𝒂𝒘 𝒕𝒉𝒊𝒔 𝒕𝒚𝒑𝒆 𝒐𝒇 𝒕𝒆𝒔𝒕𝒊𝒎𝒐𝒏𝒚 𝒕𝒘𝒐 𝒘𝒆𝒆𝒌𝒔 𝒋𝒐𝒊𝒏𝒆𝒅 𝒍𝒊𝒏𝒌. 𝑴𝒓. 𝑮𝒆𝒓𝒂𝒓𝒅𝒐 𝑰 𝒊𝒏𝒗𝒆𝒔𝒕𝒆𝒅, 𝑰 𝒈𝒐𝒕 𝒕𝒆𝒏 𝒕𝒊𝒎𝒆𝒔 𝒑𝒓𝒐𝒇� | 22:00 |
T42 | 𝒓𝒔. 𝒈𝒖𝒚𝒔 𝒕𝒉𝒊𝒔 𝒊𝒔 𝒓𝒆𝒂𝒍 𝒏𝒐𝒘 𝑰 𝒄𝒂𝒏 𝒑𝒂𝒚 𝒎𝒚 𝒃𝒊𝒍𝒍𝒔 𝒘𝒉𝒊𝒍𝒆 𝑰 𝒘𝒐𝒓𝒌 𝒇𝒓𝒐𝒎 𝒉𝒐𝒎𝒆. 𝒀𝒐𝒖 𝒄𝒂𝒏 𝒋𝒐𝒊𝒏 𝒕𝒉𝒊𝒔 𝒍𝒊𝒏𝒌 𝒋𝒐𝒊𝒏 𝒂𝒈𝒐 𝒂𝒏𝒅 𝑰 𝒓𝒋𝒊. 𝒉𝒆𝒍𝒑𝒆𝒅 𝒎𝒆 𝒐𝒖𝒕 𝒘𝒉𝒆𝒏 https://t.me/joinchat/AAAAAE-uQttmd | 22:00 |
T42 | Helen %lastname% was added by: Helen %lastname% | 23:23 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!