Thursday, 2019-08-15

T42熊頭 %lastname% was added by: 熊頭 %lastname%02:00
T42<DylanVanAssche> @adampigg the IIO adapter needs some work, however, it's a bit difficult if none of the sensors are showing up in Messwerk atm :D06:47
spiiroin@adampigg occasional source of problems: constantly reporting sensor input devices get misinterpreted as user activity -> display does not blank07:11
spiiroin"evdev_trace -t" and "mce -Tq -l event-input.c:evin_iomon_device_add --auto-exit=0" might be useful for quick check07:12
T42<adampigg> spiiroin: thx07:17
T42<adampigg> @eugenio_g7: wanna send a new tag for -config- ?07:17
T42<Thelkishere> SailfishOS supports pubg?08:02
T42<austenite> @Thelkishere this group is for people porting sailfish, you can ask here
T42<adampigg> mal: so, options for scaling are the merged commit from eugenio, or fixing the scaling in the kernel?  It wasnt clear to me yesterday what you were referring to :)08:28
Mister_Magistermal: any idea for nothing provides needed by droid-config?08:39
mal@adampigg yes, those are the options09:11
T42<adampigg> mal: this is the errror we have
T42<adampigg> updating dcd submodule didnt help09:18
Mister_Magistermal: do you know maybe how to fix camera button?09:21
mal@adampigg the problem is probably sparse/usr/libexec/droid-hybris/vendor/lib64/librecovery_updater_msm.so09:24
malyou probably need to somehow ignore the dependencies of that09:25
Mister_Magistermal: thanks, also any idea about camera button?09:26
mal@adampigg something similar to "%define __requires_exclude ^.*$" used in droid-hal-device, not sure what exactly but maybe some search tells how to use that09:27
malMister_Magister: button should just work if the event code is correct09:27
malMister_Magister: meaning show evdev_trace -i09:27
Mister_Magisterbut it doesn't just work09:27
* Mister_Magister needs to download evdev_trace somehow09:27
Mister_Magistergimme second09:27
Mister_Magistercnf is for that :P09:29
Mister_Magisterthe problem is that fxtec doesn't have wifi working :P09:31
Mister_Magisterand i don't have any 3.1 device09:31
T42<adampigg> wifi does work, just not for u :D09:57
T42<adampigg> mal: well, i did put it in there as a hack ... it should really be installed with dhd anyway10:04
ghosalmartinso am tryning to modularise my wlan, it used to work on hybris-14.1 but on 15.1 it just builds as part of the kernel, missing dep?10:13
r0kk3rzmissing kernel flag sounds like10:15
malghosalmartin: is CONFIG_MODULES enabled?10:15
ghosalmartinmal: ahh10:15
ghosalmartintime to find out if theyve broken the modular build10:16
ghosalmartinwell now .config is correct :D10:17
ghosalmartinaandd yes htey have10:21
ghosalmartinhow important is CONFIG_NETFILTER_TPROXY does anyone know?11:00
ghosalmartinit provies transparent proxy support11:00
ghosalmartinor is ipv6 used at all :P11:02
ghosalmartinmodular support of that is borked11:03
ghosalmartinI guess lets see how systemd deals with no ipv611:05
malcan't that be builtin?11:25
ghosalmartinmal: doesnt seem so, if i modularise the wlan ipv6 seems like it has to as well11:39
ghosalmartinmal: if I build dhd and then install it to device will it flash any new kernels?11:39
malghosalmartin: droid-hal install will flash kernel if you have the needed kernel flashing scripts done11:40
ghosalmartinmal: hmm i dont think i do, well i didnt include them so if their no there by default11:41
malghosalmartin: most parts are but device specific info is needed in config sparse11:41
malghosalmartin: and
ghosalmartinmal: well wifi works now11:47
ghosalmartinlets see about bt :P11:47
malghosalmartin: are you using bluebinder for bt?11:53
malghosalmartin: that is recommended for 15.111:54
malassuming of course that the device is using binderized bluetooth11:54
ghosalmartinmal: i dont have init.svc.bluetooth :(11:57
ghosalmartinbut now I have a module called bluetooth_power12:01
ghosalmartinhmm wonder if that will mean the hciattach might work now...12:01
malghosalmartin: ok maybe you need to use normal bluez way12:01
malghosalmartin: wondering why bluetooth is not binderized12:02
malthere is the wrapper binder service available for using old hal via binder12:02
malghosalmartin: for example fp2 lineage 16.0 has binderized bluetooth12:03
ghosalmartinlemme just try the hciattach, this module might of been the missing piece12:03
malghosalmartin: did that device have rome chip?12:04
ghosalmartinmal yep12:07
malghosalmartin: if the hciattach doesn't help maybe you could try to build the default wrapper binder service and try to use it12:09
malandroid.hardware.bluetooth@1.0-impl and android.hardware.bluetooth@1.0-service also maybe libbt-vendor12:10
ghosalmartinif I get hciattach working i could just go back to using hybris-14.1 and avoid tis graphics bug12:13
malghosalmartin: but so far rome or cherokee chips have been a pain12:13
malghosalmartin: wondering how mainline developers use those because mainline kernel has drivers for some rome chip(s)12:14
ghosalmartinmal: i honestly don't know tbh. I tried a while back with mainline but i didnt have a clue what i was doing12:18
ghosalmartini should of just checked the manifest to see if it was binderized12:19
ghosalmartinand lineage-16.0 doesnt have it either12:19
malghosalmartin: binderizing those in compulsory but can make things easier12:22
ghosalmartinmal: is there an example for the default wrapper binder service?12:24
ghosalmartini assume it wasnt done because the devs didnt care :P12:25
ghosalmartini am tempted to give up on this device and move onto marlin12:25
malghosalmartin: afaik it's just about building the packages I mentioned
ghosalmartinfair enough ill give it a shot thanks :)12:31
malghosalmartin: sorry12:31
malmissed one part12:31
malactually the whole entry
ghosalmartinmal: well its in there, i assume a hybris-hal will build it/12:35
malghosalmartin: not sure, if not then just manually build the packages12:36
ghosalmartinyeah always can do12:39
malghosalmartin: not sure how that manifest will be handled12:41
malor do you already that bluetooth entry there?12:41
ghosalmartinmal: nah I dont, i had to add it in12:42
ghosalmartinbtw mal dunno if this matters for the bluebinder caused a reboot after it failed waiting12:53
malghosalmartin: did you add the needed kernel config12:55
malforgot to mention that12:55
ghosalmartinmal: thats enabled12:56
ghosalmartinit still caused the phone to reboot12:56
malok :(12:58
ghosalmartinmal: well ive built the stuff I need to manually12:59
malghosalmartin: not sure what the manifest entry does13:00
malif it's needed in the other end13:00
malI mean in /system13:00
ghosalmartinmal: well ill copy the bins and lib13:02
ghosalmartinand see what happens13:02
ghosalmartini need to figure out a way to sync my two amchines in terms of file13:03
ghosalmartinmaybe just rsync13:03
ghosalmartinmal: yeah /system has a manifest.xml13:09
ghosalmartinmal this was already there
ghosalmartinbut i cant find the service ondevice :(13:13
malghosalmartin: that is missing the service13:13
ghosalmartinmal: i mean its already part of the lineage source, but i cant find it on evice in /system13:15
malghosalmartin: like I said, that is not building the actual service, just the lower level thing13:16
ghosalmartinhmm okay ive built the service :P13:17
ghosalmartinwell now init.svc.bluetooth-1-0 is running13:26
malhopefully it works13:27
ghosalmartintimed out :(13:28
malanything in logcat or dmesg?13:28
ghosalmartinmal: yeahhh buddy, its attempted to interact with it and failed. thats good13:29
malghosalmartin: /system/bin/hwservicemanager: getTransport: Cannot find entry android.hardware.bluetooth@1.0::IBluetoothHci/default in either framework or device manifest.13:33
ghosalmartinyeah thats fine i think thats not a massive issue13:33
malghosalmartin: so the manifest is not correct, of something generated from it13:33
ghosalmartini have more logs though13:33
malI think that manifest entry and whatever is generated from it are needed13:34
ghosalmartinI think I need to fix the rfkill stuff13:34
malghosalmartin: what does rfkill list?13:34
ghosalmartinID TYPE      DEVICE        SOFT      HARD13:34
ghosalmartin 0 wlan      phy0     unblocked unblocked13:34
ghosalmartin 1 bluetooth bt_power unblocked unblocked13:34
ghosalmartinill give that a shot13:37
ghosalmartinmal: any idea how i can disable the kernel building strictly?13:47
malghosalmartin: what do you mean?13:49
ghosalmartinits complaining about unused variables and i just wanna build13:52
ghosalmartini think I need to reflash anyways cause ive made a mess with bluez13:52
ghosalmartinam gonna head home and clean up the wifi work first then do the next bit13:52
ghosalmartinbut atleast we have initing of the chipset, thanks for the help :D13:53
ghosalmartinfor anyone wondering my earlier issue if CONFIG_IPV6 breaking was due to it being modularised in the codebase somewhere, a quick CONFIG_IPV6=y fixed the issue15:48
malghosalmartin: good16:02
ghosalmartinmal: realistically if i want bt does that mean am going to have to manually build the bt service and libbtvendor? or is there somewhere i can add it in my device repo so it can occur in the standard hybris-hal?16:03
ghosalmartinhas this been seen before? /usr/sbin/sensorfwd: symbol lookup error: /usr/lib/sensord-qt5/ undefined symbol: _ZN6Config13configurationEv16:42
ghosalmartinmaybe it needs libhybris hook?16:42
malghosalmartin: you had old middleware on OBS17:32
malghosalmartin: is there a reason why those were old, I updated most17:32
malghosalmartin: is mce-plugin-libhybris intentionally old, sometimes it is17:33
malghosalmartin: I updated all middleware in your bullhead OBS project, it will build a moment and then you should update17:42
Mister_Magistermal: key is reportingh KEY_CAMERA and KEY_CAMERA_FOCUS17:53
malMister_Magister: do you see the events when you press camera button?17:55
malghosalmartin: updated packages have now been built, feel free to update17:55
Mister_Magistermal: maybe some config is needed?17:57
malshould not be needed18:00
malMister_Magister: are the key codes reported normal?18:02
Mister_Magistermal: the ones i wrote above are the ones reported18:02
Mister_Magisteri will give you log if you wanty18:02
Mister_Magisteralso mal i installed 3.1 sfos on 5z (the device with broken gui)18:03
malMister_Magister: I meant the actual key code values18:04
maljust show some output from evdev_trace -t18:05
Mister_Magisterah i will18:05
Mister_Magistergood thing test_hwc2 isn't crashing but isn't working either :P18:05
Mister_Magisterwell giving logs from device with not working networking isn't easy18:07
malI only want the keycode raw values18:08
Mister_Magisterthe 0x210/KEY_CAMERA_FOCUS?18:08
maland the other also18:09
Mister_Magisterand the0x0d4 KEY_CAMERA18:09
malhmm, that looks odd18:09
Mister_Magisterif it's wrong can we change them in kernel18:14
malMister_Magister: well that is correct for kernel but looks like android usually uses 0x2fe (KEY_CAMERA_SNAPSHOT) and that is mapped in
malfor reference
Mister_Magistercan we cahnge it then?18:20
Mister_Magisterfocus doesn't seem to be working either18:21
malwondering if focus is used or not18:21
Mister_Magisterhard to chech18:22
Mister_Magisterwell when its dark and i touch button it lights up led so i think it is18:23
malMister_Magister: maybe you could add your camera keycode to that droid.kmap and regenerate the keymap file (or just rebuild config package) and see if that helps18:27
malMister_Magister: I mean mapping also 0x210 as Camera18:27
maloops I mean 0xd418:28
malso adding "keycode 212 = Camera"18:29
Mister_Magistermal: but forking submodule isn't the nicest thing to do is it?18:51
Mister_Magisteroooh so if i add it configs it will work aswell18:55
malMister_Magister: read the source, afaik you can copy both files18:55
Mister_Magisterye ye thanks!18:55
Mister_Magisteri'm being dumbass again18:55
malMister_Magister: note that you need to copy both18:55
Mister_Magisterboth files?18:55
malMister_Magister: both kmap files in the configs folder18:55
Mister_Magisterye ye18:56
T42<adampigg> @eugenio_g7 Shouls scale be 0.0001 or 0.0000001 ?18:56
malMister_Magister: so copy the whole folder to config repo main folder and then edit the droid.kmap18:56
Mister_Magisteryes yes i get it18:56
* Mister_Magister mal is not used to someone to get stuff18:56
Mister_Magisteror "better be safe than sorry"18:57
Mister_Magistermal: nope not working19:49
Mister_Magistermal: i think i actually need to replace it19:51
malMister_Magister: i doubt it, maybe you need to check with some test app if you can get the key to Qt level19:59
malMister_Magister: or even just add debug printing to camera qml if there is some key handling in those20:00
Mister_Magisterwait i might not updated droid-configs20:05
Mister_Magistermal: ye it's working i just didn't update configs20:18
malMister_Magister: ok20:21
Mister_Magistermal: can we add it to common?20:22
malmight be useful, make a PR20:22
Mister_Magisterah sorrey20:22
Mister_Magisterdone o720:26
malMister_Magister: you could mention in the description that 212 is in linux kernel header
maljust to show that the keycode is not some non-standard code but a normal code20:30
malMister_Magister: I was thinking of nitpicking about order of the keycodes but then I noticed there is one that is not in numerical order20:32
Mister_Magistermal: any idea why would usb networking not work?20:35
Mister_Magisteralso mal couple days ago there was guy with internal partitinos in storage in settings same thing here, is there fix for that?20:36
malMister_Magister: might need some tinkering of udisks2 whitelist?20:42
Mister_Magistermost likely20:42
Mister_Magisteras settings probably consiedered sd* to be external/sdcard20:42
Mister_Magistermeanwhile on efi everything is sd*20:42
malsd* is not very common in android devices20:43
malat least it didn't use to be20:43
Mister_Magistermal: because it's only in efi20:43
Mister_Magistersnapdragon 835 845 855 are using efi20:43
Mister_Magisterxbl insted of sbl20:43
Mister_Magisterit's new thing ofcourse not many have it :P20:44
malthat explains it then20:44
Mister_Magisterbut there gonna be more from now on20:45
malsure, then it needs to be fixed20:45
Mister_Magisterso how do i blacklist/fix it?20:46
Mister_Magisteralso number of partitions differs20:46
Mister_Magisterexternal partitions start at sdh for me20:46
Mister_Magisterand microsd card will be most likely mmcblk020:47
malwhat do you mean by external partitions?20:48
Mister_Magistersorrey i mean external devices20:48
Mister_Magisterlike pendrive for example20:48
Mister_Magisterlbt: can you patternise fxtec:t5? please20:49
Mister_MagisterCould not configure supporting library. hmm21:10
Mister_Magisterwhere are the logs21:10
Mister_Magisteroh lol i was disconnected21:17
Mister_Magistermal: do you have any experience with fixing video playback on android9?21:18
malMister_Magister: it should work, are mini services running?21:33
Mister_Magisteryes they are21:33
malhow do logs look like?21:35
malwhole logs21:35
malMister_Magister: I thought GST_DEBUG=droid*:5 is the correct way?21:37
Mister_Magisterwhatever is correct way?21:37
malalso I meant dmesg and logcat which might tell if something else is wrong21:37
Mister_Magistermeh i can't ssh to it nor connect over usb21:38
Mister_Magistermal: gst debug
Mister_Magisternothing in dmesg only falling services in loop21:40
Mister_Magistermal: logcat
malfailing services can be important21:46
malMister_Magister: you have so much failing like "E QMI_FW  : QMUXD: WARNING qmi_qmux_if_pwr_up_init failed!" qmi is quite important for many things21:47
Mister_Magisterwhat can i say xd21:48
malwell you need to fix those21:48
malMister_Magister: is that device using separate vendor partition?21:48
malthat do you have in straggler files?21:49
Mister_Magisteryes like i think ever treble device (it's the fxtec)21:49
Mister_Magisterif its using efi aka 835 it has to have vendor21:49
Mister_Magistermal: what files21:49
malMister_Magister: in droid-hal spec the straggler files define21:50
maldid you add such a thing, show what you have in out/..../root/21:50
Mister_Magister i don't think so21:50

Generated by 2.17.1 by Marius Gedminas - find it at!