| mal | from /data/.stowaways/sailfishos | 01:27 |
|---|---|---|
| T42 | <elros34> @xx1s1 unless your device is system-as-root and you did not revert skip_initramfs commit from kernel then it will boot android from /system | 06:09 |
| mighty17[m] | i cannot ssh or telnet again now | 06:16 |
| mighty17[m] | https://paste.debian.net/1209375/ | 06:16 |
| mighty17[m] | and ip is different now https://paste.debian.net/1209376/ | 06:20 |
| T42 | <XAP2P> gt-p3110? (re @SailfishFreenodeIRCBridgeBot: <mighty17[m]>https:/...) | 06:47 |
| mighty17[m] | p3100 | 06:47 |
| mighty17[m] | but will work on p3110 as well | 06:47 |
| T42 | <XAP2P> oHhhh (re @SailfishFreenodeIRCBridgeBot: <mighty17[m]>p3100) | 06:48 |
| T42 | <XAP2P> i have 3113(3110, but red), hard bricked(( | 06:48 |
| mighty17[m] | too many people have this tab :P | 06:48 |
| T42 | <XAP2P> i have 3113(3110, but red), hard bricked :(( (edited) | 06:48 |
| mighty17[m] | emmc dead? | 06:48 |
| T42 | <XAP2P> Нихуя се : https://irc.thaodan.de/.imgstore/748328b0/file_2558.jpg | 06:49 |
| mighty17[m] | 3113 is 3110 with ir blaster | 06:49 |
| T42 | <XAP2P> ok :D (re @SailfishFreenodeIRCBridgeBot: <mighty17[m]>3113 is...) | 06:49 |
| T42 | <XAP2P> Mine red | 06:49 |
| mighty17[m] | thats hard bricked? | 06:49 |
| T42 | <XAP2P> i updated recovery from recovery, and brick (re @SailfishFreenodeIRCBridgeBot: <mighty17[m]>emmc de...) | 06:49 |
| mighty17[m] | can u go to download mode? (power + vol down) | 06:49 |
| T42 | <XAP2P> No download mode, only charging | 06:50 |
| mighty17[m] | let it charge for sometime (like an hour or two) and then unplug it and try | 06:50 |
| mighty17[m] | as when it is charging it will not go in download/recovery mode | 06:50 |
| T42 | <XAP2P> and I also have a Chinese cable, from which it is very rarely detected by a computer | 06:51 |
| mighty17[m] | oof, if it detects, we can flash proper twrp recovery via download mode and fix it | 06:51 |
| T42 | <XAP2P> hmm, but I saw once that it was defined as a processor (omap 4340 or other) (re @XAP2P: and I also have a Ch...) | 06:51 |
| mighty17[m] | yup it will show as omap4430 in dmesg if u connect to pc when it doesnt have any charge | 06:52 |
| T42 | <XAP2P> hmm (re @SailfishFreenodeIRCBridgeBot: <mighty17[m]>yup it ...) | 06:53 |
| T42 | <XAP2P> Okay | 06:53 |
| T42 | <XAP2P> I will try if-that thanks a lot for the info. | 06:54 |
| mighty17[m] | btw the red looks cool xD | 06:54 |
| T42 | <XAP2P> yep (re @SailfishFreenodeIRCBridgeBot: <mighty17[m]>btw the...) | 06:55 |
| T42 | <XAP2P> only the lid is easily erased, a little gray in the middle | 06:55 |
| mighty17[m] | its been 9 years lol | 06:55 |
| T42 | <XAP2P> ohhhh (re @SailfishFreenodeIRCBridgeBot: <mighty17[m]>its bee...) | 06:56 |
| T42 | <XAP2P> https://irc.thaodan.de/.imgstore/77caeea6/file_2559.png | 06:56 |
| mighty17[m] | but still i like to make roms for it lmao | 06:58 |
| xx1s1 | Thanks for the answers, I'll probably just restrat the build process, see if I missed anything | 06:58 |
| T42 | <XAP2P> ok, I will test if I find it, or you give XD (re @SailfishFreenodeIRCBridgeBot: <mighty17[m]>but sti...) | 07:02 |
| mighty17[m] | i did not understand what do u mean to say? | 07:03 |
| T42 | <XAP2P> if there is a sfos, I will install it :D (re @SailfishFreenodeIRCBridgeBot: <mighty17[m]>i did n...) | 07:08 |
| T42 | <XAP2P> btw, stuoud google translate | 07:08 |
| T42 | <XAP2P> btw, stupid google translate (edited) | 07:08 |
| mighty17[m] | guess what, there is a sfos (that too made by me lol) | 07:17 |
| mighty17[m] | https://forum.xda-developers.com/t/gnu-linux-3-4-0-24-sailfish-os-for-samsung-galaxy-tab-2.4230841/ | 07:17 |
| T42 | <XAP2P> :о (re @SailfishFreenodeIRCBridgeBot: <mighty17[m]>https:/...) | 07:21 |
| T42 | <XAP2P> THX | 07:21 |
| T42 | <eriki73> @duhjokersinhell , I get only lol and rndis | 07:23 |
| T42 | <eriki73> @duhjokersinhell , I get only lo and rndis (edited) | 07:23 |
| T42 | <duhjokersinhell> does wlan.ko come up when you do lsmod | 07:25 |
| T42 | <eriki73> Yes, it did | 07:27 |
| T42 | <duhjokersinhell> hmmm ok did you push the new module to /vendor/lib/modules | 07:28 |
| T42 | <eriki73> It is in /lib/modules . I'd prefer not to touch the vendor partition... | 07:29 |
| T42 | <BigChadCat> run insmod /lib/modules/wlan.ko then (re @eriki73: It is in /lib/module...) | 07:30 |
| T42 | <BigChadCat> android usually uses such from vendor | 07:31 |
| T42 | <duhjokersinhell> i dont think its gonna work there naturally android 9 with pixel 3a reads the modules from /vendor/lib/modules. | 07:31 |
| T42 | <BigChadCat> agreed | 07:31 |
| T42 | <eriki73> @BigChadCat , the module is loaded as can be seen from lsmod and /sys/modules | 07:31 |
| T42 | <BigChadCat> idk much about sfos, can help you with the android-related part | 07:32 |
| T42 | <eriki73> It is loaded by systemd-modules-load | 07:32 |
| T42 | <edp_17> 'Morning all! I've spotted a dim red light glowing next to the front camera on my S2 (i9100). I'm not sure what sensor that is (light or proximity) but on Android, it only lights up (still very dim) when I wake up the screen. However, on sfos it is always on. How can (if it is even possible) to not have it always on? (Can it burn out or something?) | 07:35 |
| T42 | <duhjokersinhell> right but your still loading it from the android parts and android reads the modules from vendor. if you really dont want to touch vendor you may be able to symlink it but even when i compiled my nethunter kernel for p3a for UT i had to build the module and push it to vendor with adb push to get the wifi to work | 07:35 |
| T42 | <eriki73> Well I guess I can bind mount it... @duhjokersinhell | 07:36 |
| T42 | <duhjokersinhell> i dont know much about sailfish os yet. question does it incorporate system and vendor | 07:37 |
| T42 | <eriki73> @duhjokersinhell , sfos installs in userdata and boot... | 07:37 |
| T42 | <duhjokersinhell> i know but do you make a system image | 07:38 |
| T42 | <eriki73> @duhjokersinhell , Only boot.img and tar file for rootfs... | 07:39 |
| T42 | <duhjokersinhell> where is wlan.ko currently located again | 07:41 |
| T42 | <edp_17> It's confirmed, that is the proximity sensor next to the camera and emits very dim red light. (re @edp_17: 'Morning all! I've s...) | 07:41 |
| T42 | <eriki73> @duhjokersinhell , /lib/modules/*/wlan.ko now I made a bindmount to /vendor/lib/modules/wlan.ko | 07:46 |
| T42 | <duhjokersinhell> ok now try ifconfig -a | 07:47 |
| mighty17[m] | you're welcome XAP2P | 07:48 |
| T42 | <eriki73> @duhjokersinhell , https://paste.ubuntu.com/p/DG4gjG7zMr/ | 07:48 |
| T42 | <eriki73> Nothing... | 07:48 |
| T42 | <TheVancedGamer> Try adding insmod to .rc files and then push wlan.ko to vendor/lib/modules (re @eriki73: Nothing...) | 07:50 |
| T42 | <eriki73> @TheVancedGamer , Android already inserts it automatically... | 07:51 |
| T42 | <TheVancedGamer> Is wlan built as module in stock kernel? (re @eriki73: @TheVancedGamer , An...) | 07:51 |
| T42 | <eriki73> lsmod shows it is loader. | 07:51 |
| T42 | <eriki73> @TheVancedGamer , yes... | 07:51 |
| T42 | <TheVancedGamer> Hm... | 07:51 |
| T42 | <eriki73> *loaded | 07:51 |
| T42 | <TheVancedGamer> This sounds confusing | 07:51 |
| T42 | <eriki73> According to Alfred it worked ootb on Halium... | 07:52 |
| T42 | <TheVancedGamer> Pixel 3a? | 07:52 |
| T42 | <eriki73> Yes... | 07:52 |
| T42 | <TheVancedGamer> hm | 07:52 |
| T42 | <duhjokersinhell> i hate tell you but that module has to go to vendor. see system and vendor work hand in hand and android hardware and software read those modules from vendor. when you change kernel stuff for the pixel 3a it really screws around with stuff your best bet is to push the new module to vendor | 07:52 |
| T42 | <TheVancedGamer> I said that too (re @duhjokersinhell: i hate tell you but ...) | 07:53 |
| T42 | <TheVancedGamer> But isn't wlan already built as a module? | 07:53 |
| T42 | <eriki73> So, bind mounting isn't enough? | 07:53 |
| T42 | <duhjokersinhell> stock aosp yes and buiding it inside the kernel tree doesnt work well | 07:53 |
| T42 | <TheVancedGamer> Nope... | 07:53 |
| T42 | <duhjokersinhell> nope | 07:53 |
| T42 | <TheVancedGamer> Build wlan as module in kernel and then push wlan.ko into vendor (re @eriki73: So, bind mounting is...) | 07:54 |
| T42 | <TheVancedGamer> Or install a vendor with wlan module | 07:54 |
| T42 | <duhjokersinhell> ok did you try this change to the configs | 07:56 |
| T42 | <duhjokersinhell> https://pastebin.com/eeNPr4uS | 07:56 |
| T42 | <duhjokersinhell> if so the kernel should load the old modules | 07:56 |
| T42 | <eriki73> Now, it is in vendor and loader from there but no wifi, @duhjokersinhell & @TheVancedGamer | 08:01 |
| T42 | <TheVancedGamer> Try adding insmod to init.rc (re @eriki73: Now, it is in vendor...) | 08:02 |
| T42 | <TheVancedGamer> insmod wlan.ko | 08:02 |
| T42 | <duhjokersinhell> have you rebooted by chance | 08:03 |
| T42 | <eriki73> @duhjokersinhell , I have rebooted after every change... | 08:04 |
| T42 | <eriki73> @TheVancedGamer , https://paste.ubuntu.com/p/kXd82fp79T/ | 08:04 |
| T42 | <duhjokersinhell> by the way you can build all those modules as built-in all except wlan.ko | 08:06 |
| T42 | <eriki73> I do... | 08:07 |
| T42 | <TheVancedGamer> Actually that's the .cfg (re @eriki73: I do...) | 08:07 |
| T42 | <TheVancedGamer> Where's the .sh that executes it? | 08:08 |
| T42 | <eriki73> In the paste, it is shown that it is running, (getprop line) | 08:08 |
| T42 | <TheVancedGamer> Where is it though | 08:08 |
| T42 | <eriki73> @TheVancedGamer , https://paste.ubuntu.com/p/hxH595sZfg/ | 08:10 |
| T42 | <eriki73> @TheVancedGamer , IRC people don't like replying so don't do it... | 08:10 |
| T42 | <TheVancedGamer> Oh | 08:10 |
| T42 | <TheVancedGamer> @eriki73 add insmod to .rc files | 08:11 |
| T42 | <TheVancedGamer> .sh might be executed too late | 08:11 |
| T42 | <duhjokersinhell> ok last idea try to build the kernel with the changes i noted in the paste to your defconfig. if that doesnt work then theres a major problem here. i know that @fredldotme builds his wlan in the tree but by experience its buggy that way i was working with someone else having a problem with wlan as well | 08:11 |
| T42 | <duhjokersinhell> also which WCNSS_qcom_cfg.ini are you using | 08:12 |
| T42 | <duhjokersinhell> https://irc.thaodan.de/.imgstore/c90a16b0/WCNSS_qcom_cfg.ini | 08:12 |
| T42 | <duhjokersinhell> try that one it seemed to help | 08:12 |
| T42 | <eriki73> @duhjokersinhell , building the kernel now | 08:16 |
| T42 | <eriki73> @duhjokersinhell , nothing... | 08:30 |
| T42 | <eriki73> neither one did help... | 08:30 |
| T42 | <duhjokersinhell> ok take a break and wait for @fredldotme to get up and around he is the maintainer for UT on pixel 3a. hes really helpful and has helped me in the past i know he was on here earlier trying to help maybe after he sees what all has been tried he can come up with something. | 08:35 |
| T42 | <duhjokersinhell> tbh this is really weird and ive had problems with the pixel 3a's wlan but so far its all been fixable and those are all the steps ive taken im hoping he can help you | 08:35 |
| T42 | <duhjokersinhell> im sorry i couldnt be more help. but id like to be a tester for you when your ready | 08:35 |
| T42 | <eriki73> @duhjokersinhell , I am using a modified version of his kernel he did have some commits on qca_cld... | 08:36 |
| T42 | <duhjokersinhell> yea hes been trying to get that all working better. i think the problem is in the way the wlan is packaged with the source it makes it hard to pinpoint the innaccuracies | 08:38 |
| T42 | <eriki73> Btw, how to get ofono work through binder? | 09:09 |
| T42 | <eriki73> It complains that binder is not a transport method... | 09:09 |
| rinigus | I am starting to work on enabling /home encryption on boot and settings app. as far as I could see in Jolla Settings, calls are done via DBus to daemons that are changing settings. | 09:12 |
| rinigus | from this it looks to be reasonable to write a lib which will deal with the tasks involved in encryption setup and later use that lib from GUI directly (on boot) and through a daemon that would be available via dbus (dbus activated?) | 09:13 |
| rinigus | as for tasks, it looks to be communication with LUKS via libcryptsetup and formatting created partition. | 09:14 |
| rinigus | any good available qt libs for that? | 09:14 |
| rinigus | does the approach make sense? | 09:15 |
| T42 | <eriki73> @duhjokersinhell , /sys/kernel/wifi/wlan/driver_version does not exist... It should and does so in UT+Droidian. Any idea why? | 09:56 |
| T42 | <eugenio_g7> rinigus: puzzling, angelfish seems to work fine even without padding here... do you have some problematic flatpak that I could try? | 10:47 |
| T42 | <eugenio_g7> oh, of course I don't have the maliit extension | 10:47 |
| T42 | <XAP2P> :D (re @SailfishFreenodeIRCBridgeBot: <mighty17[m]>you're ...) | 10:52 |
| T42 | <eriki73> @duhjokersinhell , @TheVancedGamer https://paste.ubuntu.com/p/pQrn7xDdzW/ | 11:03 |
| T42 | <eriki73> This is from Droidian dmesg grepped with wlan... | 11:03 |
| T42 | <eriki73> https://paste.ubuntu.com/p/ybZVZYf6Bq/ | 11:04 |
| T42 | <eriki73> This is the same from SFOS | 11:04 |
| T42 | <eriki73> It looks like the driver doesnt probe devices... | 11:05 |
| T42 | <eriki73> Also, should ueventd run? | 11:22 |
| T42 | <eriki73> I guess it should not.. | 11:26 |
| rinigus | @eugenio_g7: let me check without maliit extension. but for you, angelfish is with HW acceleration? | 11:26 |
| T42 | <eugenio_g7> yes | 11:37 |
| rinigus | @eugenio_g7: nope, still same issues. on sony tama/aosp10/aarch64 based - just a sec will paste details | 11:43 |
| T42 | <eugenio_g7> with tls-padding as well? | 11:43 |
| rinigus | @eugenio_g7: nope, haven't tried that. | 11:43 |
| T42 | <eugenio_g7> okay | 11:44 |
| rinigus | this is a starting command as generated by flatpak-runner: https://dpaste.org/LaeX | 11:44 |
| rinigus | (ignore linebreaks) | 11:45 |
| T42 | <eugenio_g7> tried for now with aarch64/halium9, will also try on an halium10 port (which incidentally is a sony device as well) | 11:45 |
| T42 | <eugenio_g7> ok | 11:45 |
| T42 | <eugenio_g7> this is mine: https://dpaste.org/7Qk0 | 11:46 |
| rinigus | @eugenio_g7: on aosp9 all was fine (well with 32-bit version) | 11:46 |
| rinigus | this is what I get in angelfish: https://pasteboard.co/KhYiQqE.png | 11:47 |
| rinigus | notice that webview starts not from the top (shifted bit lower). on the bottom, you see no toolbar but only blue line. no hw acceleration | 11:48 |
| rinigus | @eugenio_g7 ^ | 11:48 |
| T42 | <eugenio_g7> ok, we had issues running webkitgtk apps on aarch64 (regardless of base), armhf was fine... padding the tls area helped | 11:48 |
| T42 | <eugenio_g7> rinigus: this is how it behaves here: https://twitter.com/eugenio_g7/status/1431581365712166914 (re @SailfishFreenodeIRCBridgeBot: <rinigus>this is wha...) | 11:49 |
| rinigus | .... patiently waiting for aosp10 run .... :) | 11:50 |
| T42 | <eugenio_g7> :) turning on the device right now! :D | 11:50 |
| rinigus | @eugenio_g7: for me, with simplified `flatpak run` (no flatpak-runner) it even doesn't start using FLATPAK_GL_DRIVERS="host" - fails to create OpenGL context | 12:11 |
| rinigus | so, I was on software gl drivers with aosp10 | 12:12 |
| T42 | <eriki73> @adampigg , it seems that you had the same problem with wifi in June 2019 on some device... Did you fix it and how? | 12:34 |
| rinigus | @eugenio_g7: adding --env=LD_PRELOAD=/usr/lib/aarch64-linux-gnu/GL/host/lib64/libtls-padding.so to `flatpak run` didn't help either - still cannot create opengl context | 12:34 |
| T42 | <eugenio_g7> rinigus: ok downloaded all the runtimes and tested :D Works on halium10 aarch64 as well without padding | 12:59 |
| rinigus | @eugenio_g7: I presume you had to add more directories to be exported as well (filesystem). | 13:00 |
| T42 | <eugenio_g7> yes /odm and /apex in this case | 13:01 |
| rinigus | padding didn't help me, unfortunately. and you have HW acceleration working just fine? | 13:01 |
| T42 | <eugenio_g7> have you tried setting LD_PRELOAD outside the flatpak environment (i.e. by not using --env)? | 13:02 |
| T42 | <eugenio_g7> yes hw acceleration works fine | 13:02 |
| rinigus | @eugenio_g7: nope. can try | 13:02 |
| rinigus | @eugenio_g7: no help. neither when using flatpak as it is nor when using flatpak-runner. I guess bad luck for me, excellent for you :) | 13:05 |
| T42 | <eugenio_g7> too bad :/ If I have any other ideas I'll ping you :) | 13:06 |
| rinigus | tbh, I have spent so much time on that flatpak that I don't have great enthusiasm going into it. better work on getting encryption done. | 13:06 |
| rinigus | but maybe someone else wishes to jump on it and fix it for SFOS/aosp10. mal was curious at a time, but that was a while ago | 13:07 |
| T42 | <eugenio_g7> so it simply fails? Or is the stack still corrupted like https://github.com/sailfishos-flatpak/main/issues/21#issuecomment-835865226 | 13:10 |
| T42 | <adampigg> Big ask for me to think about that right now....... (re @eriki73: @adampigg , it seems...) | 13:10 |
| T42 | <adampigg> https://irc.thaodan.de/.imgstore/e60df422/file_2562.jpg | 13:10 |
| T42 | <adampigg> At a festival 😂 | 13:11 |
| T42 | <eugenio_g7> haha | 13:11 |
| T42 | <eugenio_g7> enjoy :) | 13:11 |
| T42 | <eriki73> Ha, sorry for disturbing... | 13:11 |
| mal | rinigus: fix what? | 13:11 |
| T42 | <adampigg> Totally fine :) (re @eriki73: Ha, sorry for distur...) | 13:11 |
| rinigus | mal: flatpak for aosp10/aarch64 | 13:12 |
| T42 | <eriki73> Ping me if/when you have time... @adampigg | 13:12 |
| T42 | <adampigg> Will do....remind me monday :) | 13:13 |
| rinigus | @eugenio_g7: it fails as https://dpaste.org/nYzB . as for stack corruption, I haven't looked | 13:15 |
| rinigus | I'll better go for a run ... | 13:16 |
| T42 | <eugenio_g7> super stupid question, have you checked if the synced libEGL libs are correct? Perhaps there are some broken symlinks? (For example here the libEGL soname is libEGL.so.1.1.0, so I had to edit the sync script) | 13:18 |
| T42 | <adampigg> Enjoy the weekend 😂 : https://irc.thaodan.de/.imgstore/d4676df8/file_2563.jpg | 13:48 |
| T42 | <eugenio_g7> hehe, cheers, just stopped raining here lol | 13:49 |
| T42 | <adampigg> Tbh, too hot and sunny here, im a burner 😂 | 13:50 |
| rinigus | @eugenio_g7: no broken links here, unfortunately. but I better focus on encryption for now. iff I get curious about flatpak on sfos/tama, now I know who to ping | 14:07 |
| T42 | <eugenio_g7> rinigus: ok, because seeing that output it looks like that the hybris stuff doesn't get started at all | 14:10 |
| T42 | <eugenio_g7> anyway if I ever happen to port sailfish to my aosp10 device I'll take a look as well :) | 14:10 |
| rinigus | @eugenio_g7: maybe, although I think I looked into hybris logs while debugging it in April. but its a while ago | 14:11 |
| rinigus | @eugenio_g7: help will be welcomed - so make the port :) | 14:13 |
| T42 | <eugenio_g7> ;) | 14:15 |
| rinigus | @eugenio_g7: note that I used our regular hybris. So, reading up on the issue and looking at your hybris (https://github.com/droidian/libhybris/blob/bullseye/hybris/egl/egl.c#L166), I could see that you have eglGetPlatformDisplay defined | 15:30 |
| rinigus | which is absent at https://github.com/libhybris/libhybris/blob/master/hybris/egl/egl.c | 15:30 |
| T42 | <eugenio_g7> yes, eglCreatePlatformWindowSurface() as well, but I thought you tried with the glvnd branch as well back then? | 15:32 |
| T42 | <eugenio_g7> which implement those as well | 15:32 |
| rinigus | @eugenio_g7: nope - I forgot all about it :) . and as it is my daily driver, will keep it away from that. | 15:33 |
| T42 | <eugenio_g7> hehe - was reading https://github.com/sailfishos-flatpak/main/issues/21 which is where I landed the other day while debugging my (unrelated to flatpak) issue | 15:34 |
| rinigus | writing it up for future me or someone else when that person will get there | 15:34 |
| T42 | <eugenio_g7> I actually think that the missing eglCreatePlatformWindowSurface and eglGetPlatformDisplay are unrelated from the issue at hand, I mean newer qt expects those but I guess it also does on armhf? | 15:40 |
| rinigus | @eugenio_g7: maybe. I haven't tested it with newer Qt on armhf - the latest flatpak platforms are not available there | 15:41 |
| T42 | <eugenio_g7> oh I see - makes sense | 15:41 |
| T42 | <elros34> yeah on armhf ubuntu or debian (can really remember: qt5.15 or something) you need either eglGetPlatformDisplay or rebuild qt with libhybris installed | 16:03 |
| T42 | <elros34> @eriki73 you probably did notice it but on your droidian logs you have v5.3.1.1T.2 driver but on sfos v6.0.0.0A | 16:06 |
| T42 | <eriki73> @elros34 , that is strange indeed. The Droidian logs are not mine but I guess it doesn't matter as the kernel is the same... | 16:07 |
| T42 | <eriki73> https://github.com/fredldotme/android_kernel_google_bonito/commit/e98fdd2693c2818cbd2de1c72e8b36b9abb786a5 | 16:14 |
| T42 | <eriki73> This happened after I forked for Droidian. This commit is on my SFOS kernel, @elros34 | 16:14 |
| T42 | <elros34> I wouldn't be so sure it doesn't matter. Wlan is usually least difficult to get working, just build as module and modprobe. I know you said it but I hope you do not forget in your tests to write /sys/kernel/boot_wlan/boot_wlan just like in droidian logs. | 16:18 |
| T42 | <eriki73> @elros34 , on SFOS I build WiFi as module | 16:19 |
| T42 | <eriki73> This makes boot_wlan not appear on the system ... | 16:26 |
| T42 | <elros34> right, just saying it's worth to try known working wlan driver version | 16:27 |
| T42 | <eriki73> The newer is known to work on UT | 16:28 |
| T42 | <elros34> ok | 16:28 |
| T42 | <eriki73> But I guess I can try... | 16:28 |
| T42 | <elros34> @edp_17 I have read your rfkill log, try this naive patch and report whether it change anything: https://pastebin.com/PjgXH5eE | 16:53 |
| T42 | <eriki73> @elros34 , older driver did not help... | 17:08 |
| T42 | <elros34> ok | 17:12 |
| T42 | <elros34> Looking at bluetooth-rfkill-event code I wonder whether all xperia x users which has busybox killall installed (default) have issue with disabling bluetooth | 17:14 |
| T42 | <elros34> yeah looks like: https://forum.sailfishos.org/t/3-4-0-onwards-bluetooth-turning-on-doesnt-work/3072/28. If there is any xperia x user here then install psmisc-tools to fix it. | 17:17 |
| T42 | <elros34> @edp_17 That would also explain why bluetooth worked for you in 3.4 with that ugly udev killall rule | 17:19 |
| T42 | <elros34> @edp_17 updated, you can ignore rfkill_name changes: https://pastebin.com/AfGxH6Wb | 18:05 |
| T42 | <edp_17> Thanks @elros34 ! I'll try these and report. | 18:29 |
| T42 | <eriki73> Got wifi, thanks to erfan. It was a vendor loading some stuff only on zygote-start... | 20:37 |
| T42 | <edp_17> It is interesting as some mentions XA2 in the comments but, luckily, I never experienced this neither on my XA2 nor on any other ports, only on the S2 (i9100). (re @elros34: yeah looks like: htt...) | 20:42 |
| T42 | <elros34> I do not think xa2 use bluetooth-rfkill-event | 20:47 |
| T42 | <edp_17> I see. Should I apply both, the naive fix and this latter? | 20:54 |
| T42 | <elros34> obviosly the latter only | 20:57 |
| T42 | <edp_17> Okay, thanks. | 21:06 |
| T42 | <eriki73> @duhjokersinhell , as you were interested of the pixel 3A, I'll ping you. I got wifi, modem, and audio even in calls working with the help of erfan. | 21:55 |
| T42 | <eriki73> Tbh, erfan did all the thinking... | 21:56 |
| T42 | <edp_17> @elros34 : I got: 'error: corrupt patch at line 126' It is moaning about line 'INFO("BT power driver unblocked");' | 22:21 |
| T42 | <edp_17> I applied that section manualy but still got errors so I applied all of them manually. Now, I rebuild bluetooth-rfkill-event and try on device. Will let you know the outcome. | 22:22 |
| T42 | <edp_17> @elros34 : I got this error when building: https://paste.ubuntu.com/p/smT45GMwnY/ | 22:26 |
| T42 | <edp_17> I think this "patcher_started = false;" should be "patcher_started = FALSE;" | 22:26 |
| T42 | <edp_17> Once I changed to uppercase it built. | 22:26 |
| T42 | <edp_17> @elros34 : I've installed both .rpm files (bluetooth-rfkill-event-1.0.9+*.rpm and bluetooth-rfkill-event-configs-mer-1.0.9+*.rpm) but BT doesn't turn on. Logcat: https://paste.ubuntu.com/p/NvgmZBTJp7/ | 22:38 |
| T42 | <edp_17> journal: https://paste.ubuntu.com/p/TybTbfMYW8/ | 22:38 |
| T42 | <edp_17> Into /etc/sysconfig/bluetooth-rfkill-event-hciattach, I have added: RFKILLNAME="bcm4330 Bluetooth" and restarted bluetooth-rfkill-event but BT still cannot be tuned on. Journal: https://paste.ubuntu.com/p/VYVqg8gSrC/ | 22:44 |
| T42 | <edp_17> In /etc/sysconfig/bluetooth-rfkill-event-hciattach I've changed RFKILLNAME="bcm4330 Bluetooth" to RFKILLNAME="bcm4330". Still the same, BT cannot be turned on. | 22:49 |
| T42 | <elros34> probably I made some mistake, you could just skip that part about rfkill-name, it's irrelevant for your current issue.. I will prepare patch with only patcher_started changes but not sure when | 22:49 |
| T42 | <edp_17> Okay, I'll remove #define DEFAULT_BCM_RFKILL_NAME "bcm43xx Bluetooth" and put back #define BCM_RFKILL_NAME "bcm4330 Bluetooth\n" | 22:52 |
| T42 | <edp_17> Is it the only one that was relevant to rfkill-name? | 22:52 |
| T42 | <elros34> no, sorry for confusion. I will create new | 22:53 |
| T42 | <edp_17> Oh, Okay. I'll wait then. Thanks! | 22:54 |
| T42 | <elros34> @edp_17 https://github.com/elros34/bluetooth-rfkill-event | 23:06 |
| T42 | <edp_17> @elros34 : Thanks, I'm trying it now. | 23:07 |
| T42 | <fredldotme> Nice stuff! Have you tested call volume yet? (re @eriki73: @duhjokersinhell , a...) | 23:10 |
| T42 | <edp_17> @elros34 : I built and installed both .rpm files, then rebooted. The device started with BT on. I switched BT off then tried to turn it on again a few times (wait a little among attempts) but no luck. BT cannot be turned on. Journal: https://paste.ubuntu.com/p/hJwzkm32qq/ | 23:16 |
| T42 | <edp_17> Oh, I also changed the BCM_RFKILL_NAME to "bcm4330 Bluetooth\n" in bluetooth_rfkill_event.c | 23:17 |
| T42 | <edp_17> I've checked and brcm_patchram_plus was running all the time. | 23:19 |
| T42 | <elros34> I probably wrongly understand how it works | 23:23 |
| T42 | <edp_17> I've tried this new code with that ugly udev patch but that doesn't work. | 23:30 |
| T42 | <elros34> yeah that will not help with this changes. I have force pushed changes, please test again. | 23:38 |
| T42 | <edp_17> Building... | 23:41 |
| T42 | <edp_17> Almost. Device started with BT off. I could turn it on then off in Settings. (brcm_patchram_plus stopped running when BT turned off). However, when I tried to turn it on again, it started to drop that timeout messages. brcm_patchram_plus was started and then never stopped even if I toggled the BT in settings a few times: https://paste.ubuntu.com/p/TVHzZz5BDs/ | 23:49 |
| T42 | <edp_17> Once I killed brcm_patchram_plus, I could turn BT on again in Settings. | 23:49 |
| T42 | <edp_17> Okay, here it is how it "works" now: | 23:53 |
| T42 | <edp_17> 1. tap on Bluetooth in Settings (BT turns on) | 23:53 |
| T42 | <edp_17> 2. tap on Bluetooth in Settings (BT turns off) | 23:53 |
| T42 | <edp_17> 3. tap on Bluetooth in Settings (blinking) | 23:53 |
| T42 | <edp_17> 4. kill brcm_patchram_plus | 23:53 |
| T42 | <edp_17> 5. tap on Bluetooth in Settings (blinking) | 23:53 |
| T42 | <edp_17> 6. kill brcm_patchram_plus | 23:53 |
| T42 | <edp_17> 7. tap on Bluetooth in Settings (BT turns on) | 23:53 |
| T42 | <edp_17> At least bluetooth-rfkill-event.service doesn't crash. | 23:53 |
Generated by irclog2html.py 2.17.1 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!
