**** BEGIN LOGGING AT Wed Mar 06 02:59:57 2019
Mar 06 05:38:21 <ubptgbot> <JBBgameich> @Daniel, Thanks. I'll apply it in the afternoon.
Mar 06 08:14:48 <ubptgbot> <Neptvne> Just a question after using halium install to flash rootfs and system image is enough or should I mount the rootfs.img and system.img in adb shell
Mar 06 08:28:13 <ubptgbot> pzzzt was added by: pzzzt
Mar 06 09:54:04 <ubptgbot> <NotKit> @Neptvne, should be enough + you need to flash halium-boot
Mar 06 12:28:09 <ubptgbot> <Neptvne> @NotKit, Yes I've flashed it, just thought I need to mount both rootfs and system.img since it's not booting.
Mar 06 14:10:04 <ubptgbot> <JBBgameich> @JBBgameich, done
Mar 06 14:10:05 <ubptgbot> <austenite> hello , succesfully built hybris boot and system image, plasma is working and accessible through ssh but blank screen on device, in wifi test i ran this command 'echo sta > /sys/module/wlan/parameters/fwpath' , it doesnt give any output
Mar 06 14:10:22 <ubptgbot> <austenite> [Edit] hello , succesfully built hybris boot and system image for my device garlic, plasma is working and accessible through ssh but blank screen on device, in wifi test i ran this command 'echo sta > /sys/module/wlan/parameters/fwpath' , it doesnt give any output
Mar 06 14:10:48 <ubptgbot> <JBBgameich> @austenite, The wifi command isn't supposed to give any output, but you might be able to connect to the network using nmcli or nmtui.
Mar 06 14:11:05 <ubptgbot> <austenite> @JBBgameich, thanks , i will try
Mar 06 14:11:30 <ubptgbot> <JBBgameich> About graphics: Are you using the caf or generic image?
Mar 06 14:11:40 <ubptgbot> <austenite> @JBBgameich, caf
Mar 06 14:46:00 <ubptgbot> RegiSlash was added by: RegiSlash
Mar 06 20:33:43 <ubptgbot> <dohniks> Fwd from Nebrassy: Seems like I can't solve this, I have no idea what caused this to start happening, it was working fine before, also it isn't the router's Mac but the phone's Mac with the second character being wrong
Mar 06 22:26:54 <ubptgbot> <NotKit> (Photo, 960x1280) https://irc.ubports.com/6dk2EBMi.png test_hwc2 running on MTK Android 8 device based on phh-treble
Mar 06 22:27:17 <ubptgbot> <NotKit> still lots of crashing services in LXC and not minimized build, unfortunately
Mar 06 22:48:32 <ubptgbot> <phhusson> @NotKit, nice. what's your adherence to modified kernel? if i get your image and replace lxc with chroot will it just work? :D
Mar 06 22:49:19 <ubptgbot> <NotKit> running Linux in chroot or Android init in chroot?
Mar 06 22:49:41 <ubptgbot> <phhusson> well atm you're running android init in lxc no? i was thinking android init in chroot
Mar 06 22:49:52 <ubptgbot> <phhusson> though yeah linux in chroot could work i guess
Mar 06 22:49:54 <ubptgbot> <NotKit> Android init in LXC, yes
Mar 06 22:50:13 <ubptgbot> <phhusson> @NotKit you're still using libhybris and relying on android's surfaceflingre right?
Mar 06 22:50:37 <ubptgbot> <NotKit> using libhybris, but not relying on SurfaceFlinger, just HWComposer (2) API
Mar 06 22:51:09 <ubptgbot> <phhusson> uhhhh. you mean hwcomposer2 treble hal?
Mar 06 22:52:51 <ubptgbot> <NotKit> yes on that device (normally hwcomposer1 HAL was used)
Mar 06 22:52:59 <ubptgbot> <NotKit> [Edit] yes on that device (normally hwcomposer1 HAL was used before Treble)
Mar 06 22:53:13 <ubptgbot> <phhusson> why are you using libhybris then?
Mar 06 22:53:24 <ubptgbot> <phhusson> you managed to build hidl libs with glibc no?
Mar 06 22:54:41 <ubptgbot> <NotKit> that's @mariogrip. the code can be adapted to use that (for me linux-hidl is still crashy, unfortunately), but then we still rely on libEGL/libGLESv2, so no practical sense to do it for the moment
Mar 06 22:54:57 <ubptgbot> <phhusson> i see
Mar 06 22:54:58 <ubptgbot> <phhusson> cool
Mar 06 22:57:20 <ubptgbot> <NotKit> actually, I diffed kernel config, shouldn't Treble devices ship with namespaces enabled as that is used in Android now?
Mar 06 22:57:41 <ubptgbot> <phhusson> mount namespaces are needed since android 5 i think
Mar 06 22:57:47 <ubptgbot> <phhusson> but no, other namespaces are not used
Mar 06 22:58:36 <ubptgbot> <phhusson> https://github.com/LineageOS/android_device_xiaomi_msm8953-common/commit/7d80fbb988526ef81a6436d02ef2ef137697d219 such commits still make me crazy.... how the hell is that device-specific...
Mar 06 23:14:11 <ubptgbot> <NotKit> @phhusson, reminds me how MediaTek devices used to have different battery units, and... oh yeah, CAF structs in Android 7.1 headers
Mar 06 23:14:51 <ubptgbot> <phhusson> arf I didn't mean to past this here actually
Mar 06 23:15:12 <ubptgbot> <phhusson> @NotKit don't worry such pains still exist with treble... though usually it doesn't make device unusable
Mar 06 23:15:26 <ubptgbot> <phhusson> like devices not reporting battery voltage, devices not reporting at signal level
Mar 06 23:15:50 <ubptgbot> <phhusson> at least, ABI is never broken
Mar 06 23:15:52 <ubptgbot> <phhusson> that's something
Mar 06 23:16:34 <ubptgbot> <phhusson> no struct offsets to reverse-engineer or what. also, since most HALs are used in java, reverse-engineering the interface is really super easy
Mar 06 23:16:55 <ubptgbot> <phhusson> (I reversed some in https://github.com/phhusson/vendor_interfaces if that's of any interest)
Mar 06 23:21:12 <ubptgbot> <NotKit> that should be helpful, thanks (though probably still quite a while before we can care about these)
Mar 06 23:36:36 <ubptgbot> <minlexx> do you know anything about why qualcomm device tree config forbids writing time to rtc (/dev/rtc0) ? and why there is a proprietary TimeService for that ?
Mar 06 23:41:13 <ubptgbot> <phhusson> @minlexx that's not what timeservice does. timeservices stores in /data the difference between the "rtc" and real time
Mar 06 23:41:26 <ubptgbot> <phhusson> that means their rtc simply can't be set at all
Mar 06 23:41:58 <ubptgbot> <minlexx> @phhusson, Oh, I see. Thanks
**** ENDING LOGGING AT Thu Mar 07 02:59:57 2019