| Umeaboy | If anyone is awake I seem to have trouble getting a certain .rc from the mounted system/vendor directory. The file is in the wrong place. | 00:40 |
|---|---|---|
| T42 | <CSmartly> please do VPN android apps like one for ProtonVPN work on latest SailfishOS? | 00:42 |
| Umeaboy | CSmartly: Do you have Android support enabled? | 00:52 |
| Umeaboy | Android support as standard in Sailfish OS is NOT free just so you know. You have to purchase a license to enable that possibility. | 00:53 |
| T42 | <CSmartly> @Umeaboy [CSmartly: Do you have Android support enabled?], I have | 00:54 |
| Umeaboy | Well, then I guess you should find the app in any store then. | 00:55 |
| T42 | <CSmartly> yes app is there, but I am asking whether it works, many apps historicly didnt work and among them were VPNs, so I had to set up with files directly in system of SailfishOS | 00:56 |
| Umeaboy | Try and see. :) | 00:56 |
| Umeaboy | Then you'll know. | 00:56 |
| Umeaboy | F-droid have it. :) | 00:56 |
| T42 | <CSmartly> no, its not for me F-droid has what, working free VPN? | 00:57 |
| Umeaboy | The app ProtonVPN that you asked for. | 00:57 |
| T42 | <CSmartly> I have it installed yes, not sure if it work as other didnt I tried in the past. Didnt want to spend time making account etc if its still not working | 00:58 |
| T42 | <CSmartly> I will leave it for the person who gets the phone to figure it out | 00:58 |
| Umeaboy | I don't have Android app support as I don't need it so I just checked in my LOS 18.0 phone if F-droid had it. | 01:00 |
| T42 | <CSmartly> yes but rememeber if it is in the store doenst mean it works on Sailsfis | 01:00 |
| T42 | <CSmartly> some apps dont work, because it is not native Sailfish app android app doesnt get same access etc | 01:01 |
| Umeaboy | Fair enough. | 01:03 |
| Umeaboy | What the heck...... I can't have extract-files.sh find init.rilmptcp.rc in the right place no matter how I change proprietary-files.txt | 01:04 |
| Umeaboy | I have tried everything. | 01:04 |
| Umeaboy | make hybris-hal fails to find that .rc file. | 01:04 |
| T42 | <birdzhang> @pketo [birdzhang[m]: is it 64bit port], yep | 01:09 |
| Umeaboy | Hi birdzhang! :) | 01:11 |
| Umeaboy | How are you? | 01:11 |
| Umeaboy | Sorry. Forgot the @ :) | 01:11 |
| Umeaboy | It's late here. | 01:11 |
| T42 | <birdzhang> Hi | 01:11 |
| Umeaboy | I was wondering if you can help me get this line correct..... init.rilmptcp.rc:system/etc/init/init.rilmptcp.rc | 01:12 |
| Umeaboy | It's not there. | 01:12 |
| Umeaboy | It's in the main dir instead. | 01:12 |
| Umeaboy | I tried lots of changes, but I still can't find it. | 01:12 |
| T42 | <birdzhang> can you show your device tree github link? | 01:15 |
| Umeaboy | @birdzhang: https://github.com/whatawurst/android_device_samsung_beyond2lte | 01:17 |
| Umeaboy | system/etc/init/init.rilmptcp.rc: file not found in source because that file is in the system partition when you open it. | 01:21 |
| Umeaboy | I mounted the system.img to system/ and the vendor.img to /system/vendor/ just like the extract proprietary blobs page for Lineage suggests so I don't think I made a boo boo. | 01:23 |
| Umeaboy | I am double checking and writing every step down for remembering if and when I start over. | 01:23 |
| T42 | <birdzhang> sorry, i'm not good at this😅 | 01:24 |
| Umeaboy | No problem. | 01:26 |
| T42 | Brent %lastname% was added by: Brent %lastname% | 04:55 |
| pketo | @birdzhang added | 05:00 |
| T42 | <birdzhang> Thank you | 05:39 |
| T42 | supirlelik96 was added by: supirlelik96 | 09:03 |
| T42 | <birdzhang> @adampigg How do you fixed audio on the volla 64 port? I have no audio on ginkgo(Redmi Note8) aarch64 too | 13:01 |
| T42 | <adampigg> I was missing a symlink to libandroidicu.so (64bit) | 13:03 |
| T42 | <Verevka86> @birdzhang [@adampigg How do you fixed audio on the volla …], https://github.com/SailfishOS-sagit/droid-config-sagit/commit/5a277fff3b79b9ce6970f51050328c86c12bfe6a | 13:04 |
| T42 | <Verevka86> Check this | 13:04 |
| T42 | <birdzhang> I'm on hybris-16.0, thank you all, let me try | 13:04 |
| T42 | <supirlelik96> https://github.com/mer-hybris/hadk-faq#android-base-specific-fixes | 17:45 |
| T42 | <supirlelik96> I need to run Android base specific fixes for hybris17. 1 los17 | 17:45 |
| T42 | <supirlelik96> What to do if there are no Android base specific fixes for hybris 17.1? | 17:45 |
| mal | https://github.com/mer-hybris/hadk-faq/pull/8/files | 17:52 |
| T42 | <supirlelik96> @mal [https://github.com/mer-hybris/hadk-faq/pull/8/ …], Thank you | 18:11 |
| T42 | <supirlelik96> @mal [https://github.com/mer-hybris/hadk-faq/pull/8/ …], Вот | 18:13 |
| piggz | mal: https://source.android.com/devices/sensors/sensors-hal2 | 18:51 |
| piggz | is there any implementatino of this interface for sfos? | 18:51 |
| mal | piggz: not yet, not sure how we should handle the event fmq | 18:52 |
| piggz | mal: volla implements that, not 1.0 | 18:53 |
| mal | piggz: yeah, I think there might have been some other device also | 18:53 |
| piggz | mal: wondering how the 10 ii doesnt use that?? | 18:54 |
| mal | good question | 18:54 |
| piggz | ill put sensors to the back of the queue then for now :D | 18:56 |
| mal | well I was the one who did the binder backend to sensorfw so maybe I should have some idea but need to see when I have enough time to think | 18:57 |
| piggz | sure, i also need your help with sailjail ;) | 18:58 |
| mal | why do I have to solve everything :) | 18:59 |
| piggz | cos u know everything | 19:00 |
| piggz | your fault being the sailor that interacts the most here :D | 19:00 |
| T42 | Ajdj %lastname% was added by: Ajdj %lastname% | 19:01 |
| piggz | by all means, fire me some tips for debugging .. i tried strace already | 19:01 |
| piggz | actually, strace fails becuase sailjail is suid | 19:13 |
| piggz | and that cant be fixed https://superuser.com/questions/247888/strange-strace-and-setuid-behaviour-permission-denied-under-strace-but-not-run | 19:14 |
| mal | hmm | 19:16 |
| piggz | so, im willing to debug, just unsure how ;) | 19:21 |
| T42 | <elros34> add -vv to sailjail :P | 19:21 |
| piggz | well, it tells me there is an issue witha qopengl context that doesnt happen outside of sailjail ... https://paste.mozilla.org/c6roGCyq | 19:28 |
| mal | piggz: wondering if there is some device node it can't access or something | 19:31 |
| piggz | mal: logcat for with/without sailjail shows identical output | 19:35 |
| mal | hmm | 19:35 |
| piggz | mal: something extra in dmesg when it fails though.... | 19:37 |
| piggz | working: | 19:38 |
| piggz | [83025.428744] (5)[1664:jolla-calculato][ION]ion_handle_get_by_id_nolock: can't get handle by id:0 | 19:38 |
| piggz | [83025.428787] (5)[1664:jolla-calculato][ION]ion_ioctl:ION_IOC_FREE handle is error | 19:38 |
| piggz | not working: | 19:38 |
| piggz | [82980.921640] (6)[1316:jolla-calculato][ION]ion_handle_get_by_id_nolock: can't get handle by id:0 | 19:38 |
| piggz | [82980.921802] (6)[1316:jolla-calculato][ION]ion_ioctl:ION_IOC_FREE handle is error | 19:38 |
| piggz | [82980.982400] (4)[1459:QSGRenderThread]dump_emit: __kernel_write fail: -32 | 19:38 |
| piggz | [82980.996613] (5)[1316:jolla-calculato][ION]ion_client_destroy:hdl=00000000f8e1b928,buf=00000000bb609417,ref=2,kmp=0,clnt=1316,dbg=gralloc | 19:38 |
| piggz | [82980.996670] (5)[1316:jolla-calculato][ION]ion_client_destroy:hdl=000000008937c2da,buf=000000003040c9dc,ref=2,kmp=0,clnt=1316,dbg=gralloc | 19:38 |
| piggz | [82980.996689] (5)[1316:jolla-calculato][ION]ion_client_destroy:hdl=00000000e7a29a93,buf=00000000f68ccade,ref=2,kmp=0,clnt=1316,dbg=gralloc | 19:38 |
| mal | hmm, -32 seems EPIPE | 19:40 |
| T42 | <NotKit> is there a way to test you can access /dev/mali*, /dev/ion, /dev/pmsg inside firejail? | 20:43 |
| T42 | <NotKit> that should be enough for GPU driver | 20:43 |
| T42 | <NotKit> maybe also /dev/mtk_disp_mgr | 20:43 |
| mal | yeah, maybe changing the sailjail permissions manually could help figuring out why it fails | 20:48 |
| T42 | <NotKit> btw, regarding sensors hal 2 and fmq, I experimented with old Marius linux-hidl project to compile Android HIDL-related C++ parts natively | 20:49 |
| T42 | <NotKit> but it seems overkill to use it just for that | 20:50 |
| mal | yeah, need to check a bit how the queue stuff is implemented to see if we can easily make a linux version of it | 20:51 |
| mal | piggz: can you grab a strace of a working app run so maybe we can see if something in there matches the rules | 20:52 |
| piggz | sure... | 20:52 |
| T42 | <NotKit> https://github.com/ubports/sensorfw/commit/1f5afd4eceaef26fd0a930df7d6e8661f4d25d2d - from HAL interaction point it's something like that | 20:52 |
| piggz | mal: https://paste.opensuse.org/61019285 | 20:54 |
| mal | @NotKit at least that gives some idea what might be needed but still I would go with more native approach | 20:55 |
| piggz | it access pmsg0 | 20:55 |
| piggz | and /dev/ion | 20:56 |
| T42 | <NotKit> and /dev/mali0 | 20:56 |
| mal | I don't see anything about those in the rules | 20:57 |
| piggz | mal: is that something i need to add somewhere? | 21:06 |
| mal | not sure yet | 21:06 |
| T42 | <b100dian> I'm trying to enable double tap to wake and found a lengthy printf that enables that in android | 22:44 |
| T42 | <b100dian> basically to reach this point https://github.com/b100dian/Xiaomi_Kernel_OpenSource/blob/tucana-hybris-17.1/drivers/input/touchscreen/fts_521/fts.c#L6705 | 22:44 |
| T42 | <b100dian> I need to actually send an input_event struct like `printf "%b" '\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x00\x01\x00\x04\x00\x00\x00' > /dev/input/event4` | 22:45 |
| T42 | <b100dian> the problem is - enabling this seems to not be enough in sailfish + mce | 22:45 |
| T42 | <b100dian> I've run `mcetool --set-doubletap-wakeup=always` and `mcetool --set-low-power-mode=enabled` and even stopped mce and run it with verbose and stderr | 22:47 |
| T42 | <b100dian> (of course that one thing that doesn't work is setting those \x' | 22:55 |
| T42 | <b100dian> 00 values in the ini file.. | 22:55 |
| T42 | <elros34> have you checked whether you have any input event on double tap? | 22:58 |
| T42 | <b100dian> ok, I did see a dmesg from fts kernel, but how do I do that, just cat the event device? | 22:59 |
| T42 | <b100dian> yup | 22:59 |
| T42 | <elros34> evdev_trace to see which device is that then eved_trace -t <input device number> | 23:00 |
| T42 | <b100dian> Yes, here's both ways https://pastebin.ubuntu.com/p/rnrB497HYJ/ | 23:01 |
| T42 | <b100dian> (now I know key_wakeup was on this channel, grepping) | 23:01 |
| T42 | <elros34> maybe this is enough: https://github.com/mer-hybris/droid-config-f5121/commit/424ba6ee4d2b8700dd65abf944a249162aab9fbd | 23:05 |
| T42 | <b100dian> Yup, it works like this, thanks! | 23:08 |
| T42 | <b100dian> Are there any perils for this alias that you know? | 23:08 |
| T42 | <elros34> no idea but this f5121 is official device so.. | 23:10 |
| T42 | <b100dian> Ok:) One of the previous discussions was piggz pointing out that 'you dont want to map KEY_WAKEUP to KEY_POWER, as every time there is a ring indication, the screen will invert', but I'm not sure what this means | 23:11 |
| T42 | <elros34> wasn't that for modem triggering tat event on pine* device? | 23:11 |
| T42 | <elros34> I am only wondering whether there is better way to enable double tap like sysfs interface:/sys/class/input/event5/device/wakeup_gesture | 23:12 |
| T42 | <b100dian> I started with that.. I don't have that sysfs file. I have a bunch of others, you can configure gestures in shapes of letters to wake up or smth | 23:13 |
| T42 | <b100dian> but this is the only way I could trigger by reading driver sources.. | 23:14 |
| T42 | <b100dian> well, actually it is simpler - that can be done only once at boot time if I understand correctly. | 23:14 |
| T42 | <b100dian> of course, I need to test battery first - or fix the proximity sensor and make mce read binary strings from ini files - but hopefully it's a hardware thing that doesn't consume much.. | 23:17 |
| T42 | <elros34> You could write few lines of code in kernel driver and add sysfs interface similar to mentioned wakeup_gesture. For your other gestures I remember such a app which allows you to trigger actions based on gesture https://github.com/kimmoli/gesture-daemon | 23:18 |
| T42 | <elros34> or maybe even this one:https://github.com/sailfishos-oneplus5/onyx-gesture-settings-plugin | 23:21 |
| T42 | <b100dian> Thats actually a great idea, thanks! | 23:22 |
| T42 | <b100dian> Ni, i dont need the other gestures, just things ive read in there | 23:22 |
| T42 | <b100dian> The touchscreen can send letters or something:) | 23:22 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!