| spiiroin | @DylanVanAssche note that mce normally does not "process" volume key events, that happens at ui side. but mce does implement policy that dictates whether ui should use or ignore volume key events it gets (directly from evdev source) -> if ui actually gets events and chooses not to act on them, something at mce side might be causing policy misinterpretation | 06:50 |
|---|---|---|
| spiiroin | say, if there is music playback -> volume key policy should be "use 'em". that "there is music playback" is determined from ohm signaling - if mce misinterprets that -> ui might get told to ignore volkeys | 06:51 |
| T42 | <DylanVanAssche> spiiroin: mce seems to detect properly the events (display is off -> ignore volume key for example). | 08:36 |
| T42 | <DylanVanAssche> So it's higher up the chain where it goes wrong | 08:36 |
| spiiroin | @DylanVanAssche nb: there was some reporting in tjc where volkeys did misbehave occasionally - lipstick restart / device reboot apparently can help -> so something weird going on in general | 11:43 |
| T42 | <DylanVanAssche> spiiroin: Okay, but there's nothing else between mce and lipstick? | 11:48 |
| spiiroin | @DylanVanAssche regarding volume keys? not really. the relevant dbus sig is https://git.merproject.org/mer-core/mce-dev/blob/master/include/mce/dbus-names.h#L1058 handled around here https://git.merproject.org/mer-core/lipstick/blob/master/src/volume/volumecontrol.cpp#L77 | 12:15 |
| T42 | <DylanVanAssche> spiiroin: Thanks! Now I'm sure which thing I need to debug, I thought there was something else between mce and Lipstick for volume managment | 12:59 |
| T42 | <Zosenko> Hello, how can i get back action with swype into any apps in SailfishOS? | 13:26 |
| T42 | <elros34> What do you mean? | 13:33 |
| T42 | <Lukapanio> Cannot find '/usr/libexec/droid-hybris/system/bin/miniafservice', disabling 'miniaf': No such file or directory | 14:16 |
| T42 | Andrea7638 was added by: Andrea7638 | 14:30 |
| T42 | <Andrea7638> Vuoi guadagnare 5€ per ogni utente che inviti? Beh è semplicissimo. Serve solo un email. | 14:30 |
| T42 | <Andrea7638> - Scarica TIK TOK dal play store. | 14:30 |
| T42 | <Andrea7638> - Vai in basso a destra e | 14:30 |
| T42 | <Andrea7638> premi su "Me". | 14:30 |
| T42 | <Andrea7638> - Ti registri con l'email o il numero. | 14:30 |
| T42 | <Andrea7638> - Una volta registrato premi torna su "Me" e premi su raccogli gemme. | 14:30 |
| T42 | <Andrea7638> - Premi su codice invito e inserisci questo codice: `6621896` | 14:30 |
| T42 | <Andrea7638> Ora copia il tuo codice e invita i tuoi amici 👍. | 14:30 |
| T42 | <Andrea7638> Puoi prelevare solo una volta al giorno! AFFRETTATI SCADE OGGI. | 14:30 |
| T42 | <hacker12455> how can I route the internet from my PC to Sailfish? | 14:48 |
| vknecht | @hacker12455: hadk pdf section 9.3.3 ? | 14:54 |
| T42 | <hacker12455> don't have IP assigned to usb0 | 14:58 |
| T42 | <hacker12455> I don't even see the interface on host | 15:00 |
| T42 | <hacker12455> ok fixed that | 15:03 |
| *** Gizmummy2005 is now known as Gizmokid2005 | 15:37 | |
| rinigus | piggz || @adampigg: around? wanted to ask you about bootctl fix for pro1. | 18:22 |
| T42 | <adampigg> Sure | 18:23 |
| rinigus | adampigg: I have the same bug as you had for pro1 - bootctl doesn't mark boot as successful | 18:24 |
| rinigus | and logcat is the same as well, with ../../../../ referenced. | 18:24 |
| T42 | <adampigg> One moment... | 18:24 |
| rinigus | sounds like its gpt-utils incompatibility with symlinks as setup by sfos for storage | 18:25 |
| T42 | <adampigg> Yup | 18:25 |
| rinigus | have you managed to fix it? | 18:25 |
| piggz | yes | 18:26 |
| piggz | https://paste.mozilla.org/FccTaAxg | 18:26 |
| rinigus | excellent! have you pushed it upstream? | 18:26 |
| rinigus | piggz: and if you did, then where :) | 18:28 |
| piggz | rinigus: just in our git server which contains one single repo for the entire pro1 source ... its a nightmare to work with | 18:29 |
| rinigus | piggz: I have the same issue with sony xz2. I'll test your patch and let you know if it works. pretty sure its the same problem as I have | 18:30 |
| piggz | fingers crossed | 18:30 |
| rinigus | piggz: thanks! | 18:32 |
| vknecht | ll | 18:41 |
| vknecht | oops | 18:41 |
| piggz | rinigus: sweet | 18:43 |
| rinigus | piggz: not tested yet - will let you know as soon as done | 18:44 |
| rinigus | :) | 18:44 |
| vknecht | piggz, crazy... https://framapic.org/gallery#tGPaBGREtxsd/SBM9o57P1avR.jpg | 18:45 |
| vknecht | too bad, no width/height in pixels :-/ | 18:46 |
| vknecht | wonder if resolution (and maybe others) should be stuffed manually like at https://github.com/piggz/harbour-advanced-camera/blob/1e418434ef22b5cd379f8a8f2e227ac0c1ebe41d/qml/pages/CameraUI.qml#L117 | 19:02 |
| Thaodan | rinigus: You can add a mount unit to your sparse image to mount selinux fs | 19:10 |
| rinigus | Thaodan: you cannot. Systemd is refusing to do so, at least as provided by sfos. But from quick glance over it's sources, I expect it will refuse to do it with other configure as well | 19:13 |
| rinigus | Thaodan: unless it mounts it itself with selinux support enabled | 19:13 |
| Thaodan | what does it say? | 19:13 |
| rinigus | Thaodan: it was specific error message refusing to do so when I wrote corresponding mount. You could go systemd source for /sys/fs/selinux to get logic behind it | 19:15 |
| rinigus | Hence the script to mount it. Which works perfectly and brings all up with gui | 19:16 |
| rinigus | As I don't have that config around, can't test it either... | 19:17 |
| rinigus | Btw, does anyone have xperia 10 around and could check if selinuxfs is mounted there? | 19:18 |
| Thaodan | Is selinux enabled? Maybe systemd wants to do it on its own. | 19:20 |
| Thaodan | https://github.com/systemd/systemd/blob/e1439a1472c5f691733b8ef10e702beac2496a63/src/core/mount-setup.c#L117 | 19:24 |
| rinigus | Thaodan: its not. as far as I understood, its from current ea. that was pointed out by piggz over here few days ago. | 19:26 |
| rinigus | Thaodan: as for that part in the code, see comment above it | 19:26 |
| rinigus | mal: increasing to 2048 (instead of 2014 as in https://github.com/libhybris/libhybris/commit/c7f87904345f19e05a0b9170d9b3898e92e9a910) helped | 19:29 |
| Thaodan | I know it was explanation to link where the issue is. | 19:29 |
| rinigus | never thought I'll be happy to see tutorial screens :) | 19:29 |
| rinigus | sorry, 1024 increased to 2048 on my device | 19:29 |
| rinigus | Thaodan: yes, just wanted to be clear | 19:30 |
| rinigus | Thaodan: but, in general, if they get selinux and audit supported in systemd - this should make it simpler to debug selinux issues if any. I am still on selinux enabled setup and now will have to start debugging all 'smaller' issues. | 19:32 |
| Thaodan | Have you pushed your changes to hybris-initrd? | 19:32 |
| Thaodan | Ok I'll see if my device as similar issues, afer the whole android 10 thing.. | 19:33 |
| rinigus | Thaodan: which changes you mean? in initrd-helpers? | 19:33 |
| Thaodan | the changes to the configs handling | 19:34 |
| Thaodan | *configfs | 19:34 |
| rinigus | Thaodan: I don't know what you mean. usb rndis? that's device-specific, I think | 19:35 |
| Thaodan | I think the path changes are maybe not device specific. | 19:36 |
| rinigus | what I haven't done is to figure out intellegent way of checking for mmblk0p* or sda* in https://github.com/sailfishos/initrd-helpers/blob/master/find-mmc-bypartlabel | 19:43 |
| rinigus | bin/sh scripting gurus ^ | 19:43 |
| Thaodan | You could do [ ! -d /sys/class/block/mmcblk0p* ] || [ ! -d /sys/class/block/sda* ] | 19:51 |
| rinigus | Thaodan: thanks, somehow got frozen as soon as I saw #!/bin/sh | 19:55 |
| wiktorek140 | mal: could you look at this? https://pastebin.com/1itdXW2E | 19:56 |
| mal | rinigus: did you check how long the string actually was? | 21:00 |
| mal | wiktorek140: do you have the 15.1 specific .rc files form faq? | 21:01 |
| mal | *from | 21:01 |
| wiktorek140 | mal: yea, all mentioned in repo | 21:03 |
| wiktorek140 | i cloned repo too | 21:05 |
| mal | wiktorek140: and selinux is correctly disabled? | 21:05 |
| wiktorek140 | mal: config_audit=n is enough? or should i change commandline too? | 21:06 |
| mal | wiktorek140: you should have config_audit=y, selinux_bootparam=y and then add selinux=0 to kernel commandline | 21:08 |
| wiktorek140 | mal: oh, will change to that options RN, thanks | 21:09 |
| wiktorek140 | mal: it will be enough to flash only kernel? | 21:09 |
| mal | yes | 21:17 |
| mal | wiktorek140: check the correct name for the selinux_bootparam, can't remember the whole parameter name | 21:24 |
| mal | kernel checker does tell that | 21:24 |
| rinigus | mal: no, I have not. should do though... | 21:28 |
| wiktorek140 | mal: same errors | 21:40 |
| mal | wiktorek140: did you check the config on device "zcat /proc/config.gz | grep SELINUX" and also /proc/cmdline for the selinux=0 part | 21:42 |
| wiktorek140 | mal: both in places | 21:44 |
| mal | wiktorek140: exactly same errors? | 21:46 |
| wiktorek140 | mal: send whole journal? | 21:46 |
| mal | wiktorek140: is that device 32 or 64 bit? | 21:47 |
| wiktorek140 | mal: 32 bit | 21:47 |
| mal | do you have the correct selinux_stubs in the .rc files? | 21:47 |
| mal | default is 64 bit libs, you need to manually change those | 21:48 |
| wiktorek140 | mal: https://pastebin.com/URkGcU71 | 21:48 |
| wiktorek140 | so in *.rc files in sparse i need to change things to 32bits? | 21:48 |
| rinigus | piggz: I am going to test it as in https://gist.github.com/rinigus/3fa16dfbc533010c2e9f2fdacc7ee9d8 | 21:49 |
| rinigus | piggz: I think it should cover better corner cases where buflen is smaller than path_max | 21:50 |
| rinigus | as required by realpath | 21:50 |
| deathmist | wiktorek140: looks like so, you change "lib64" part in "setenv LD_PRELOAD /usr/libexec/droid-hybris/system/lib64/libselinux_stubs.so" lines of those .rc files to simply "lib" | 21:51 |
| wiktorek140 | deathmist: only in sparse is enough? | 21:53 |
| deathmist | you also need to do the modification to /usr/libexec/droid-hybris/system/etc/init/servicemanager.rc, maybe mal knows how to make the change permanent for your device since that file is provided by something else | 21:56 |
| mal | deathmist: wiktorek140 that servicemanager.rc is in https://github.com/mer-hybris/android_frameworks_native/commit/99bbbcab366f92b5b28da6f590e8cd05e987d819 | 21:57 |
| wiktorek140 | mal: thanks i will revert it in a minute | 21:58 |
| mal | wiktorek140: no, don't revert, fix the path | 21:58 |
| mal | so lib64 -> lib | 21:58 |
| wiktorek140 | only in out/ dir? | 21:58 |
| wiktorek140 | mal: fixed, thanks | 21:59 |
| mal | wiktorek140: for testing just change the path in all .rc files on device and check how it works, then later fix the files in sparse and that last one in frameworks/native/cmds/servicemanager/servicemanager.rc | 21:59 |
| wiktorek140 | mal: there are other way, copy file to lib64 xD | 22:00 |
| mal | wiktorek140: quite ugly way :) | 22:08 |
| wiktorek140 | mal: same kind of errors | 22:12 |
| wiktorek140 | mal: https://del.dog/jeqoludase maybe you will saw what missing | 22:28 |
| deathmist | wiktorek140: kinda not related but your log is being spammed by a "droid-hcismd-up.sh" script which shouldn't be used for 15.1 and above ports as a binder interface alternative is available and should be used w/ bluebinder MW (unless that fails to work) | 22:35 |
| r0kk3rz | anyone know who this is? https://gitlab.com/JhonRoun | 22:37 |
| wiktorek140 | deathmist: removed script, can you point me what i need to do to use bluebinder? | 22:37 |
| deathmist | wiktorek140: iirc you uncomment bluebinder in your droid-config patterns / add "- bluebinder" there, here's a snippet of my working 15.1 BT related defconfig changes https://pastebin.com/raw/n6DPPmfx | 22:39 |
| wiktorek140 | deathmist: thanks, will add it, but removing script dont help for now | 22:43 |
| mal | wiktorek140: not the same errors, you don't have servicemanagers failing anymore | 22:50 |
| mal | r0kk3rz: why do you ask? | 22:50 |
| r0kk3rz | this person is asking to join porters-ci | 22:54 |
| wiktorek140 | mal: can it be the reason "Sailfish invoker[3370]: libminisf is incompatible or missing. Can not possibly start the SurfaceFlinger service. If you're experiencing troubles with media try updating droidmedia (and/or this plugin).== hwcomposer module ==" | 22:54 |
| mal | Jul 25 20:22:57 Sailfish lipstick[5683]: [W] unknown:0 - Cannot find EGLConfig, returning null config | 22:54 |
| mal | Jul 25 20:22:57 Sailfish lipstick[5683]: [F] unknown:0 - EGL Error : Could not create the egl surface: error = 0x3001 | 22:54 |
| mal | that is the error | 22:54 |
| mal | r0kk3rz: no idea who that is | 22:55 |
| wiktorek140 | mal: some clue to fix that? maybe something in configs? | 22:58 |
| mal | r0kk3rz: looks like he has been on this channel | 22:59 |
| wiktorek140 | mal: test_hwcomposer works | 23:13 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
