**** BEGIN LOGGING AT Tue Jan 29 02:59:56 2019 Jan 29 06:10:44 @Daniel, Could you check if Qt VirtualKeyboard works for you? Jan 29 11:10:42 smnet1999 was added by: smnet1999 Jan 29 16:56:55 @UniversalSuperBox are you going to make more live porting sessions like the last one? Maybe this could become a series where you and Marius make ports with community members Jan 29 17:01:48 I think that would be neat Jan 29 17:01:54 I really enjoyed it. Jan 29 17:09:23 If you'd do the community variant, then if I don't manage to achieve anything with ubports on my device, count me in 😀 Jan 29 17:10:03 @UniversalSuperBox, Please next time turn on the microphone and turn off telegram notifications :) Jan 29 17:10:22 Btw you can use your phone with headphones to check for sound levels Jan 29 17:10:46 Or well obs has a peak for your input, you could make use of that too Jan 29 17:11:11 Or you could record a small piece via obs “record” function prior to the steam and play it back to check for issues Jan 29 17:11:27 The stream also got horribly desynced due to buffering Jan 29 17:11:32 @VeryOriginalUsername, was that a stream on ubports channel or what Jan 29 17:11:46 @UniversalSuperBox, Well you said it was a first one, so not a big deal Jan 29 17:11:52 @vanyasem, Just take that in mind Jan 29 17:12:14 I was streaming for a long time via obs in the past, so just sharing my experience Jan 29 17:12:22 I'm not sure how to fix that, though... Marius has high-bandwidth but high-latency Jan 29 17:12:31 I have nothing redeeming about my connection Jan 29 17:12:41 Well buffering is not a big deal when it comes to coding Jan 29 17:12:57 It would be crucial for real time games, but here - meh Jan 29 17:13:12 Just sort the mic issue out Jan 29 17:13:18 @vanyasem, Hey Ivan, hows the op3 port btw Jan 29 17:13:46 Some people in Supergroup got really excited about it Jan 29 17:14:15 @Flohack, Ssh works with UT now Jan 29 17:14:20 Apart from that, nothing much Jan 29 17:14:29 OK i see Jan 29 17:14:54 I don’t have that device, if I get to visit @sk8higher any time soon, I’ll give it a shot overnight for sure Jan 29 17:15:11 Ah ok didnt know that. Jan 29 17:15:33 I always use teleconsole or tmux when porting it Jan 29 17:15:48 It’s impressive I got it to boot plasma not even having it physically xD Jan 29 17:15:59 Lol cool Jan 29 17:16:08 I listed everyone who helped me along the way on the issue Jan 29 17:16:18 Like 4 different people among the way Jan 29 17:16:23 Whoever is free at that moment Jan 29 17:16:37 I feel we should soon address those differences that allow plasma to boot but refuse UT xD Jan 29 17:16:50 @Flohack, Well mainly I guess kernel flags Jan 29 17:16:56 I only ran the mer checker Jan 29 17:16:59 And not the Ubuntu one Jan 29 17:17:07 Maybe we dont need all of them on UT hehe Jan 29 17:17:36 @Flohack, Yeah, if there will be a list on what is different exactly, I might be more lucky Jan 29 17:17:54 It’s kinda hard to do a trial and failure path when you depend on someone to test it for you Jan 29 17:18:05 Anyways I gtg currently at gym Jan 29 17:18:18 Been a pleasure talking to you, long time no see Jan 29 17:27:32 yeah Jan 29 17:27:35 c u later ^^ Jan 29 17:39:25 to clarify, in fixup-mountpoints should there be only !(auto,mmc,swap) partitions, or all of the partitions? some devices have like 40 entries whereas some have like 3-5 Jan 29 17:39:58 In reality, only `/data` is really required to my knowledge. However, Mer prefers everything that is mountable is listed Jan 29 18:18:05 @UniversalSuperBox that's is because mounts are turned into systemd units for them Jan 29 18:18:16 while in Halium they are handled by bash script on boot Jan 29 18:19:01 [Edit] @UniversalSuperBox that's is because mounts are turned into systemd units for them at build time Jan 29 18:19:24 [Edit] @UniversalSuperBox that's is because mounts are turned into systemd units for them at droid-hal-device package buil time Jan 29 18:19:34 [Edit] @UniversalSuperBox that's is because mounts are turned into systemd units for them at droid-hal-device package build time Jan 29 19:08:08 update: I managed to fit ubports rootfs on my device's data part without messing with sd Jan 29 19:08:54 apparently 1,6 gigs is taken with rootfs and doing system.img magic (```adb push system.img /system/system.img && ln -s /system/system.img /data/system.img```) works Jan 29 19:09:00 [Edit] apparently 1,6 gigs is taken with rootfs and doing system.img magic ```adb push system.img /system/system.img && ln -s /system/system.img /data/system.img``` works Jan 29 19:12:18 @NotKit, Mer excites and confuses me. Jan 29 19:15:10 Guys Jan 29 19:15:19 Regarding the live porting sessions Jan 29 19:15:36 I stumbled across them today Jan 29 19:16:00 @VeryOriginalUsername, nope, system didn't get mounted at boot and hence no lxc 😅 Jan 29 19:16:08 can I somehow tell halium-boot to mount it? Jan 29 19:16:20 I really, really want to find everybody involved and give y'all one big hug😊 Jan 29 19:16:46 It really feels that it's everyday people porting stuff Jan 29 19:17:41 @VeryOriginalUsername, Where did you put it? Jan 29 19:17:55 I can mount it manually to /android/system Jan 29 19:18:14 How large is the system partition? Jan 29 19:18:18 1,2 gig I believe Jan 29 19:18:25 or something around that Jan 29 19:19:18 @VeryOriginalUsername, Or maybe, you can flash system.img to /system and use it entirely (?) Jan 29 19:19:40 that would be a cleaner solution, yes Jan 29 19:20:50 theorycraft: … -include system part in fixup-mountpoints … -make sure it's in fstab … -flash system.img to system part … -??? … -profit Jan 29 19:21:11 Problem: System needs to mount to have fstab Jan 29 19:21:27 [Edit] theorycraft: … -include system part in fixup-mountpoints (it's there) … -make sure it's in fstab … -flash system.img to system part … -??? … -profit Jan 29 19:22:11 @UniversalSuperBox, And if we modify halium init file to mount directly /system ? Jan 29 19:23:46 (Photo, 719x294) https://irc.ubports.com/DisEte0j.png Jan 29 19:24:17 You're confusing halium-boot and hybris-boot Jan 29 19:24:23 and myself for sure Jan 29 19:27:19 @VeryOriginalUsername, 😁 Jan 29 19:27:27 can't I just include /system in rootfs's /etc/fstab? Jan 29 19:27:31 or is it too late Jan 29 19:28:18 That's too late Jan 29 19:28:23 And halium-boot overwrites the fstab Jan 29 19:28:31 ouch Jan 29 19:28:58 You might need to modify the `halium` script to do what you want, then submit the changes upstream Jan 29 19:29:20 @UniversalSuperBox, If I'm not mistaken, the fstab used by the rootfs is a dummy file Jan 29 19:29:22 It's not hard, you'll need to add a new layout type and a way to switch to it. Jan 29 19:29:25 @WaseemAlkurdi, Exactly Jan 29 19:31:25 layout type = partitions mounted on boot in this case I presume? Jan 29 19:31:47 Layout type determines where the rootfs and systemimage can be found Jan 29 19:32:01 oh, yes, that's what I meant Jan 29 19:32:15 and the way to switch to it could be a variable in devicetree Jan 29 19:32:26 Yes, or a boot parameter Jan 29 19:32:28 like HALIUM_USE_ALT_SYSIMG Jan 29 19:33:17 You need to consider updating as well in that case... some additions to UT's system-image or whatever PM is planning to use Jan 29 19:33:41 @UniversalSuperBox Is there anything preventing me from using hybris-boot instead of halium-boot on A/B Pixel XL? Jan 29 19:34:00 Are you trying to boot Ubuntu Touch or not? Jan 29 19:34:06 yes Jan 29 19:34:18 oops, perhaps I wasn't the one asked Jan 29 19:34:44 lol, it was @WaseemAlkurdi Jan 29 19:34:45 @UniversalSuperBox, I am planning to Jan 29 19:34:57 The design of Ubuntu Touch is preventing you from using hybris-boot Jan 29 19:35:25 @WaseemAlkurdi, It seems PM doesn't work (yet?) with halium-boot and hybris has no a/b support Jan 29 19:35:38 Not Ubuntu Touch itself .. I just want anything to fix the damned graphics😂 Jan 29 19:35:54 @Sergobot, It's possible that upstream hybris-boot does. Would need investigation. Jan 29 19:35:59 @Sergobot This is exactly what bothers me Jan 29 19:36:10 @UniversalSuperBox, Oh that sounds promising Jan 29 19:36:14 it has to since sailfish works with xperia xa2 Jan 29 19:36:17 The Sailfish guys say that A/B is supported by hybris-boot Jan 29 19:36:20 [Edit] it has to since sailfish works on xperia xa2 Jan 29 19:36:27 EXACTLY! @Konrad Jan 29 19:36:33 Orly Jan 29 19:36:42 Gotta try it out then :P Jan 29 19:36:54 @Sergobot, Same here! Jan 29 19:37:01 @UniversalSuperBox, Can you send the link? :) Jan 29 19:37:43 Sure, I'll search the internet for it Jan 29 19:38:40 Apparently Mer hasn't moved this to their GitLab instance yet: https://github.com/mer-hybris/hybris-boot Jan 29 19:40:42 And as for why I'd prefer to use halium-boot rather than hybris-boot: https://github.com/Halium/projectmanagement/issues/55 Jan 29 19:42:45 And for the work I attempted to do a year ago to make UT compatible with hybris-boot (which failed): https://github.com/ubports/ut-halium-compat Jan 29 19:50:23 @UniversalSuperBox, Looking at the advantages of halium-boot I tend to think making it work with PM is more promising than backporting a/b stuff to hybris-boot Jan 29 19:58:53 @ilyaishere, Sure, I'll give it a shop asap. … You are on Z2580 (Asus T00f), right? (Cause that is SGX series) Jan 29 20:02:22 @Daniel, Yep Jan 29 20:22:30 either I'm blind or there's nothing about system.img neither in hybris-boot nor in halium-boot Jan 29 20:22:39 or it's unrelated and I should look somewhere else Jan 29 20:26:14 ohh, OHHHH so the image is INJECTED into rootfs, right? Jan 29 20:26:16 @Sergobot, We're not backporting anything Jan 29 20:26:27 They say that it's supported out-of-the-box Jan 29 20:26:45 Yeah, but that is hybris 8, I believe Jan 29 21:34:53 @UniversalSuperBox, surprise surprise doesn't boot no surprise in my case xd Jan 29 21:35:28 datapart expects both rootfs.img and system.img in the same partition Jan 29 21:38:51 they were both on sd Jan 29 21:39:03 Are you sure that the block device was available? Jan 29 21:39:15 this one I can't guarantee Jan 29 21:39:31 I'd really rather do the system.img on /system way Jan 29 21:41:00 Technically you could assuming there's enough room there and you manually set the fstab to use that partition Jan 29 21:41:21 wait, could /system being ro by default be the problem? Jan 29 21:41:43 I'll try giving it same mountflags as userdata Jan 29 21:43:07 In my case it was Jan 29 21:43:07 @VeryOriginalUsername Jan 29 21:43:07 Not /system per se Jan 29 21:43:07 But rootfs Jan 29 21:43:07 Init couldn't create mountpoints on a R/O rootfs Jan 29 21:43:38 (Document) https://irc.ubports.com/lX0k2La7.null Jan 29 21:43:48 I made some changes to hybris-boot to take system into account Jan 29 21:43:52 they are really dirty though Jan 29 21:45:20 @WaseemAlkurdi, how did you mount it rw? Jan 29 21:45:26 on default Jan 29 21:47:57 I know with halium-boot you can just do "touch .writable-image" in /data and it mounts rw Jan 29 21:48:22 gonna try that out rn! Jan 29 21:48:36 Good luck Jan 29 22:07:23 @VeryOriginalUsername, this doesn't seem to have any effect Jan 29 22:07:30 unless I'm doing something terribly wrong Jan 29 22:21:22 lmao, changing rootfs size to 1.75G also makes it fit, and so I managed to fit system.img on /data too Jan 29 22:27:12 oh my, the dmesg is sad Jan 29 22:41:01 @VeryOriginalUsername, but why don't you just flash system.img to system then and fix script to mount it? Jan 29 22:41:33 ln /data/system/system.img /data/system.img - a mistake here, you can not create hard links across partitions Jan 29 22:41:36 it didn't work for me either but I've tried so many dirty things that I don't know anymore Jan 29 22:41:42 @NotKit, ln -s didn't work either Jan 29 22:42:03 what do you want to boot actually? Jan 29 22:42:56 I have Gemini PDA changes for halium-boot to use "linux" partition, which could as well reside on SD card in theory Jan 29 22:45:08 ubports Jan 29 22:48:03 https://github.com/Halium/initramfs-tools-halium/blob/halium/scripts/halium#L378 - you could add some label before userdata in partlist and create ext4 partition on SD card with same label in GPT, though I don't see why just datapart=/dev/mmcblk1p1 in kernel cmdline would not work Jan 29 23:36:47 (Document) https://irc.ubports.com/PQ2gzQF7.null Jan 29 23:36:48 (Document) https://irc.ubports.com/blfxAMrY.null Jan 29 23:37:24 that's my logs, to my eye it looks like there is something wrong with cgroups/cgmanager Jan 29 23:40:59 also, any systemctl command would return ```Failed to list units: No such method 'ListUnitsFiltered'``` or ```Failed to get properties: No such interface ''``` **** ENDING LOGGING AT Wed Jan 30 02:59:57 2019